cellular mobile communication a fundamental ...

6 downloads 5005 Views 11MB Size Report
Figure Ch1.1: A typical high level illustration of cellular system. The Mobile ...... decides to send the next radio blocks (e.g., numbers 1 to 4) with CS3. During the ...... “Intuitive Guide to Principles of Communications: CDMA Tutorial”, retrieved.

CELLULAR MOBILE COMMUNICATION A FUNDAMENTAL PERSPECTIVE

Technical Note

R&D NEXTEVOLUTION

Branding Intellects

RONY. K. SAHA

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

ACKNOWLEDGEMENTS Information incorporated in this lecture note may be found fully or partly with or without any change in figures, tables, sentences, etc. in the references. The lecture note is developed only for the purpose of class room use. The lecture note developer fully acknowledges and grateful to these helpful contributions (mentioned references as well as the references there-in) to help make this lecture note developed.

Lecture Note Developer R. K. SAHA Master of Engineering (Info. & Comm. Tech.) Asian Institute of Technology, Bangkok, Thailand

2

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

PREFACE The field of Cellular Mobile Communications (CMC) is highly emerging. CMC Technologies have been in the move towards achieving ever increasing user needs for rich multimedia services. In the past less than three decades, enormous changes in CMC have been addressed in technology, standard and system level. This student handout is developed to get the students a fast but with considerable explanation of the evolution of CMC technologies, standards and systems. A good trade-off between depth-and-breath of any context is considered. The handout can be used for teaching and research both senior undergraduate students and graduate students. In the first part, fundamentals of CMC including channel propagation characteristics and modeling, multi-path fading, channel classifications and parameters, cellular concept, cellular system design fundamentals are addressed. Standards mainly Global System for Mobile Communications (GSM) and GSM Evolution and Code Division Multiple Access (CDMA) are explained in detail. In the second part (Please refer to Technical Note on Advanced Mobile Communication), we start with a general understanding of high data rate communications and the means to achieve so. Advanced enabling technologies such as Multiple-input Multiple-output (MIMO), orthogonal Frequency Division Multiplexing (OFDM), Resource scheduling, Link adaptation, Hybrid Automatic Repeat Request (HARQ) are discussed. Third generation (3G) standards and beyond 3G are addressed. Standards including Wideband CDMA (WCDMA), High Speed Packet Access (HSPA), Long Term Evolution (LTE), and LTE-Advanced (LTEAdvanced) are explained in detail. Changes in architectures from one generation to another are explained, and their requirements for changes are detailed. Students will get a seamless experience throughout the handout as they read up. There are a number of problems that are designed and collected to understand the topics discussed in the chapters. Many of these problems are solved, and others are left for the students to exercise. The purpose here is not to get the exercise solved, rather is to gain understanding how to solve. Students are therefore advised to discuss with others about the outcome of each exercise. We consider avoiding mathematical explanation as much as possible so that students do not need much mathematical background as prerequisite on mastering the matter. However, a good understanding of linear algebra, Fourier analysis, signals and systems, digital signal processing, and fundamentals of telecommunications is highly desirable. Many of the topics are taken from industry standardization body‟s whitepapers, IEEE publications, and renowned authors in the field. Please refer to the Reference section for further information. Finally, we believe that this handout will serve for understanding the fundamentals of CMC technologies, standards and systems to the reader. Should there be any further information, inquiry, or suggestion, please reach us with the followings.

Regards, R. K. SAHA

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

3

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION 4

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CONTENTS Chapter number

Title

Page

01

INTRODUCTION Fixed Wireless Networks and Mobile Wireless Networks Conventional Mobile Radio and Cellular Mobile Radio Cellular Mobile Radio Features of Cellular Radio Analog and Digital Cellular Radio Transmission Cellular Systems Cellular Radio Call Setup Procedure Cellular Radio Roaming Cellular Phone Standards First Generation Analog Systems Second Generation Digital Systems Evolution of 2G Systems Third Generation Systems

9

02

SIGNAL PROPAGATION AND PATH LOSS MODELS Radio Frequency Propagation Overview Basic Propagation Mechanisms Propagation Environments Outdoor Propagation Environments Indoor Propagation Environments Overview of Propagation Models Propagation Models and Path Loss Estimation Empirical (or statistical) models Free-Space Propagation Path Loss Model Log-Distance Path Loss Model Okumura Model Hata Model Walfisch and Bertoni Model COST-231-Walfisch-lkegami Model Site-Specific Models for Path Loss Ray-Tracing Technique 2-ray Ground Reflection Model

17

03

04

CONCEPT OF FADING IN CELLULAR RADIO SYSTEMS Multipath Fading Large-scale fading Small-scale fading Fast fading Slow fading CHANNEL CLASSIFICATION IN CELLULAR RADIO SYSTEM Classification of Channels Fading Channel Characteristics Power-Delay Profile Delay Spread First-Arrival Delay Mean Excess Delay RMS Delay Maximum Excess Delay

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

5

29

37

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Time Dispersion and Frequency-Selective Fading Coherence Bandwidth Frequency-flat and Frequency-selective fading Doppler Spread Coherence time Time-flat and time-selective Fading Channel Classes Example 05

06

CELLULAR SYSTEM DESIGN: CONCEPT & PRACTICE Cellular System Design Background and Concept Cellular Frequency Reuse Concept Conceptual Model Analytical Model Channel Assignment Strategies Handoff Strategies Practical Handoff Considerations Interference in Cellular Systems Co-channel Interference Adjacent Channel Interference Trunking and Grade of Service Improving Capacity in Cellular Systems Cell Splitting Sectoring Microcell Zone Concept An Example Cellular System Design Analytical Derivation of Frequency Reuse Factor Analytical Derivation of Capacity of Cellular Systems Traffic Channel Assignment Adjacent Channel Interference Near-End-to-Far-End Ratio

47

DIGITAL CELLULAR RADIO SYSTEMS GSM Overview GSM Services GSM Architecture Positioning GSM Telephone Types of Positioning Systems Positioning Parameters Logical Channel and Mapping GSM Speech Coding Full Rate Speech Coding Half-Rate Speech Coding Enhanced Full-Rate Speech Coding Channel Coding and Interleaving Frequency Hopping Problem GSM Signaling Procedure Registration Call Origination Paging Handover Security Authentication Encryption Key Distribution User Identity Protection GSM Transmission

76

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

6

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

GSM Receiver 07

GPRS OVERVIEW Benefits of GPRS Drawbacks of GPRS GPRS Services GPRS Architecture Transmission/signaling planes in GPRS Packet-Switched Transmission Over The Air Interface Security In GPRS Authentication in GPRS Ciphering in GPRS Security in the GPRS backbone Mobility Management in GPRS GPRS service areas Accessing the GPRS network Mobility management states Keeping track of the MS End-To-End Packet Routing PDP context activation procedure Packet switching in GPRS Data routing for a mobile MS GSM/GPRS Service Interactions Co-ordination of location area and routeing area updates Circuit-switched paging via the SGSN The special case of class-B MS Point-to-point short message service (SMS) GPRS Efficiency Limitations of GPRS

106

08

ENHANCED DATA RATES FOR GSM EVOLUTION (EDGE) Introduction GPRS and EGPRS Architecture EDGE Technology EDGE Modulation Technique Coding Schemes EGPRS Link Controlling Function Link Adaptation Incremental redundancy Packet Handling Interleaving Impact of EGPRS on existing GSM/GPRS networks EGPRS Architecture and Protocols EGPRS Benefits

128

09

EDGE EVOLUTION EDGE Evolution Performance Boost Implementing EDGE Evolution Latency Reduction Reduced TTI Faster Feedback Increased Bit Rates and Improved Efficiency Dual Carriers Higher-Order Modulation, Turbo Codes and Increased Symbol Rate Dual-Antenna Terminals Mobility Enhancements Service Coverage Summary of Edge Evolution

138

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

7

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

10

CODE DIVISION MULTIPLE ACCESS Multiple Access Technique CDMA Principle Spreding And Despreading Operation CDMA Correlation Receiver CDMA Rake Receiver Properties Of Spreading Codes M-sequences Gold Sequences Long code Short code Walsh Codes Channel Waveform Properties Forward Channel Description Pilot Channel Synch Channel Paging Channel Traffic Channel Reverse Channel 64-ary Modulation CDMA System Aspects Near/Far Problem Soft Handover Spectral Efficiency CDMA Capacity Multicell TDMA System Multicell CDMA System Practical Differences between CDMA and FDMA/TDMA Versatility with New Service Power Control Open Loop Uplink Power Control Closed Loop Uplink Power Control Downlink Power Control Mathematical Analysis Call Origination Authentication Problem CDMA 2000 Summary

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

150

8

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 01 INTRODUCTION 1.1 INTRODUCTION Fixed Wireless Networks and Mobile Wireless Networks In comparison to fixed wireless networks, mobile wireless networks differ in the followings. 

Higher loss rates due to interference (emissions of such as engines, lightning, etc.)



Restrictive regulations of frequencies (frequencies have to be coordinated. In addition, useful frequencies are almost all occupied).



Low transmission rates (Indoor up to 2 Mbit/s and outdoor up to 384 kbit/s are for pedestrian speed).



Higher delays and higher jitter (connection setup time is in the second range; several hundred milliseconds is for other wireless systems).



Lower security and simpler active attacking (radio interface is accessible for everyone and base station can be simulated result attracting calls from mobile phones).



Always shared medium (secure access mechanisms is important). 9

Conventional Mobile Radio and Cellular Mobile Radio Conventional Mobile Radio In the earliest, mobile radio system’s transmitters and receivers were designed to operate on a fixed frequency so that each radio channel was reserved to a single user. However, the total traffic for a number of frequency channels can be significantly increased if a group of channels is made available to every user in a group. This technique is called trunking. And it is essentially the same principle as the switched telephone network, where transmission facilities are made available to users on demand via exchanges. The first trunked systems were based on manual trunking, where users switched manually through different channels and listened whether or not the channel is free. In 1960s, the choice of a free channel was made automatic, i.e. mobile phones could identify a free channel from a transmitted idle tone. Another important improvement in the conventional mobile radio was the improvement in FM transmitters and receivers. When the original bandwidth of FM mobile radio telephones that was required for adequate speech quality was 120 kHz in 1930s, the required bandwidth for the same quality decreased to 60 kHz in 1950s and to 30 kHz in early 1960s. Important new feature introduced in the 1960s was also full-duplex operation instead of half-duplex operation in the mobile station. The conventional mobile radio reached technically mature level in the early 1970s by utilizing trunking, improved FM technology, direct dialing, and automatic switching. However, there was a serious capacity problem in conventional mobile radio due to the network architecture. The traditional approach to mobile radio network architecture was similar to radio or television broadcasting: a high-power transmitter on top of the highest peak in the area providing fair to average coverage over an area with radius of 60 km to 80 km.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

This meant that still a single user reserved a single radio channel over all this area for the entire duration of a call. For example, in New York City in the 1970s, this kind of system could provide 12 simultaneous mobile calls for a population of substantially over 10 million people. On uplink, although the traditional mobile radio networks utilized relatively densely spaced receiving sites, because mobile stations used low-power transmitters, sites could not always get a signal back to the middle of town. However, all these receiving sites were serving the same customer on a particular carrier frequency, and hence the use of multiple receivers did not increase the system capacity. Cellular Mobile Radio The cellular idea is to use a large number of low-power transmitters, each of them designed to serve only a small coverage area, cell, with radius of couple of kilometers. This makes it possible to reuse the same frequencies in different cells. However, due to the interference between adjacent cells, the same frequencies cannot be used in every cell, and it is necessary to skip several cells before reusing a particular frequency. Even with this restriction, the capacity increase from the broadcast-type mobile radio is impressive. Moreover, since the interference between adjacent cells is determined by the transmitter powers in these cells, the frequency reuse pattern can be easily changed by splitting a single cell to several smaller cells with lower transmitter power in the geographical areas where the traffic load is high. The consequence of small cell size is that all mobile calls cannot always be completed within the same cell in which they originated. Thus, for cellular mobile systems an operation where mobile calls are switched from one base station to another called handover (or handoff in North America) is obligatory. The cellular idea was first proposed already in 1940s. However, significant advances in technology were required before cellular networks could be practically realized. These advances include VLSI technology, microprocessors, frequency synthesizers, high capacity fast switches, etc. Features of Cellular Radio Followings are the essential features that a cellular radio system should encompass: 

In a cellular system it must be possible to route incoming calls to the mobile station (MS) regardless of its location in the operator's service area. In many cellular systems, it is also possible to route incoming calls to a MS when it is located in another operator's service area. This feature is called roaming. Cellular systems, such as NMT and GSM systems, international roaming is also possible.



It must also be possible for the MS to have access to the fixed telephone network (PSTN, Public Switched Telephone Network).



It is considered very important that ongoing calls are not interrupted regardless of MS mobility. This creates a need to reroute on-going calls, which means switching the call to another frequency channel or to another base station (BS). This procedure is called handover.



In cellular network planning, in the starting phase, for serving relatively few car users large macrocells (generally which radius > 1 km) are the most economical solution. To serve larger number of pedestrian users smaller microcells (generally which radius < 1 km) are needed to meet the capacity requirements. Indoor users can be served by either macrocells or microcells. However, crowded places such as railway stations, underground stations, warehouses indoor base stations such as picocells (generally which radius < 100 m), or femtocells (generally which radius 10 to 30 m) would be a good choice to serve very small areas. Note that the propagation effects for the desired signal and the effects of co-channel interference differ largely in macro-, micro-, pico-, and femtocells.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

10

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Analog and Digital Cellular Radio Transmission 

The second generation of cellular radio also named as digital cellular radio can provide, along with conventional speech telephony, very wide range of services such as circuit and packet switched data, short messages, facsimile, etc. Note that it is difficult to accommodate these services into analog networks; especially it is difficult to multiplex different kinds of services into one radio channel.



In addition, the digital radio transmission is also more resistant to noise and interference than analog radio transmission. Thus the radio system can be designed for lower signal-to-noise ratio (SNR) or signal-to-interference ratio (SIR) or signal-to-interference-plus-noise ratio (SINR).



The bandwidth of digital speech signal is normally higher than the bandwidth of the original analog speech signal. However, since the digital transmission is more resistant to interference, using advanced digital modulation techniques and low bit rate speech coding methods, it is possible to achieve higher spectral efficiency in digital transmission than analog transmission.



In general, most of the advantages of the use of digital technology for communications also apply for the cellular systems such as use of digital integrated circuits, error correction, equalization, performance monitorability, security, connectivity to other networks, and etc. 1.2 CELLULAR SYSTEMS

Cellular systems accommodate a large number of users over a large geographic area, within a limited frequency spectrum. High capacity is achieved by limiting the coverage of each base station transmitter to a small geographic area called cell so that the same radio channels may be reused by another base station located some distance away. A sophisticated switching technique called a handoff enables a call to proceed uninterrupted when the user moves from one cell to another. A high level illustration of cellular system is given in figure Ch1.1.

Figure Ch1.1: A typical high level illustration of cellular system. The Mobile Switching Center is responsible for connecting all mobiles to the PSTN in a cellular system. Each mobile communicates via radio with one of the base stations and may be handed off to any number of base stations throughout the duration of a call. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

11

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The mobile station contains a transceiver, an antenna, and control circuitry and may be mounted in a vehicle or used as a portable hand-held unit. The base stations consist of several transmitters and receivers, which simultaneously handle full duplex communications and generally have towers that support several transmitting and receiving antennas. The base station serves as a bridge between all mobile users in the cell and connects the simultaneous mobile calls via telephone lines or microwave links to the MSC. The MSC coordinates the activities of all of the base stations and connects the entire cellular system to the PSTN. 1 Communication between the base station and the mobiles is defined by a standard common air interface (CM) that specifies four different channels. The channels used for voice transmission from the base station to mobiles are called forward voice channels (PVC) and the channels used for voice transmission from mobiles to the base station are called reverse voice channels (RVC). The two channels responsible for initiating mobile calls are the forward control channels (FCC) and reverse control channels (RCC). Control channels are often called setup channels because they are only involved in setting up a call and moving it to an unused voice channel. Control channels transmit and receive data messages that carry call initiation and service requests, and are monitored by mobiles when they do not have a call in progress. Cellular Radio Call Setup Procedure When a cellular phone is turned on (and is not yet engaged in a call), it first scans the group of forward control channels to determine the one with the strongest signal and then monitors that control channel until the signal drops below a usable level. At this point it again scans the control channels in search of the strongest base station signal. For each cellular system, the control channels are defined and standardized over the entire geographic area covered and typically make up about 5% of the total number of channels available in the system (the other 95% are dedicated to voice and data traffic for the end-users). When a call is initiated by a PSTN telephone user

12



When a telephone call is placed to a mobile user, the MSC dispatches the request to all base stations in the cellular system.



The mobile identification number (MIN), which is the subscriber's telephone number, is then broadcast as a paging message over all of the forward control channels throughout the cellular system.



The mobile receives the paging message sent by the base station, which it monitors and responds by identifying itself over the reverse control channel.



The base station relays the acknowledgment sent by the mobile and informs the MSC of the handshake.



Then the MSC instructs the base station to move the call to an unused voice channel within the cell (typically between ten to sixty voice channels, and just one control channel are used in each cell's base station).



At this point the base station signals the mobile to change frequencies to an unused forward and reverse voice channel pair.



And another data message called an alert is transmitted over the forward voice channel to instruct the mobile telephone to ring, thereby instructing the mobile user to answer the phone.



All of these events occur within a few seconds and are not noticeable by the user.

1

Note that A typical MSC handles 100,000 cellular subscribers arid 5,000 simultaneous conversations at a time, and accommodates all billing and system maintenance functions, as well. In large cities, several MSCs are used by a single carrier.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



Once a call is in progress, the MSC adjusts the transmitted power of the mobile and changes the channel of the mobile unit and base stations in order to maintain call quality as the subscriber moves in and out of range of each base station. This is called a handoff.

When a call is initiated by a mobile phone user 

When a mobile originates a call, a call initiation request is sent on the reverse control channel.



With this request the mobile unit transmits its telephone number (MIN), electronic serial number (ESN), and the telephone number of the called party.



The mobile also transmits a station class mark (SCM) that indicates what the maximum transmitter power level is for the particular user.



The cell base station receives this data and sends it to the MSC.



The MSC validates the request and makes connection to the called party through the PSTN,



And instructs the base station and mobile user to move to an unused forward and reverse voice channel pair to allow the conversation to begin.

Cellular Radio Roaming 

All cellular systems provide a service called roaming. This allows subscribers to operate in service areas other than the one from which service is subscribed.



When a mobile enters a city or geographic area that is different from its home service area, it is registered as a roamer in the new service area. This is accomplished over the (Forward control channel) FCC since each roamer is camped on to a FCC at all times.



Every several minutes, the MSC issues a global command over each FCC in the system asking for all mobiles that are previously unregistered to report their MIN and ESN over the (Reverse control channel) RCC.



New unregistered mobiles in the system periodically report back their subscribe information upon receiving the registration request, and the MSC then uses the MIN/ESN data to request billing status from the home location register (HLR) for each roaming mobile.



Once registered, roaming mobiles are allowed to receive and place calls from that area and billing is routed automatically to the subscriber's home service provider. 1.3 CELLULAR PHONE STANDARDS

First Generation Analog Systems The main characteristics of the standards for first-generation (1G) analog cellular phones are summarized in table Ch 1.1. Systems based on these standards were widely deployed in the 1980s. The best known standard is the Advanced Mobile Phone System (AMPS) developed by Bell Labs in the 1970s and first used commercially in the US in 1983. Japan deployed the first commercial cellular phone system in 1979 with the NTT (MCS-L1) standard based on AMPS but at a higher frequency and with voice channels of slightly lower bandwidth. Europe also developed a similar standard to AMPS called the Total Access Communication System (TACS). TACS operates at a higher frequency and with lower bandwidth channels than AMPS. It was deployed in the U.K. and in other European countries as well as outside Europe. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

13

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

In addition to TACS, countries in Europe had different incompatible standards at different frequencies for analog cellular, including the Nordic Mobile Telephone (NMT) standard in Scandanavia, the Radiocom 2000 (RC2000) standard in France, and the C-450 standard in Germany and Portugal. Table Ch.1.1: First-Generation Analog Cellular Phone Standards.

Second Generation Digital Systems The main characteristics of the second-generation (2G) digital cellular phone standards are summarized in table Ch.1.2. These systems were mostly deployed in the early 1990s. Due to incompatibilities in the firstgeneration analog systems, in 1982 the Groupe Sp´ecial Mobile (GSM) was formed to develop a uniform digital cellular standard for all of Europe. In 1989 the GSM specification was finalized, and the system was launched in 1991, although availability was limited until 1992. The GSM standard uses TDMA combined with slow frequency hopping to combat out-of-cell interference. Convolutional coding and parity check codes along with interleaving are used for error correction and detection. The standard also includes an equalizer to compensate for frequency-selective fading. As the GSM standard became more global, the meaning of the acronym was changed to the Global System for Mobile Communications. In 1992 the IS-54 digital cellular standard was finalized with commercial deployment beginning in 1994. This standard uses the same channel spacing of 30 kHz as AMPS to facilitate the analog to digital transition for wireless operators along with a TDMA multiple access scheme to improve handoff and control signaling over analog FDMA. The IS-54 standard, also called the North American Digital Cellular standard was improved over time and these improvements evolved into the IS-136 standard which subsumed the original standard. Similar to the GSM standard, the IS-136 standard uses parity check codes, convolutional codes, interleaving, and equalization. Proposed by Qualcomm in the early 1990s is called IS-95 or IS-95a was finalized in 1993 and deployed commercially under the name cdmaOne in 1995. The channel chip rate is 1.2288 Mchips/s for a total spreading factor of 128 for both the uplink and downlink. The IS-95 standard includes a parity check code for error detection, as well as power control for the reverse link to avoid the near-far problem. A 3-finger RAKE receiver is also specified to provide diversity and compensate for ISI. The 2G digital cellular standard in Japan, called the Personal Digital Cellular (PDC) standard was established in 1991 and deployed in 1994. It is similar to the IS-136 standard but with 25 kHz voice channels to be compatible with the Japanese analog systems. This system operates in both the 900 MHz and 1500 MHz frequency bands.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

14

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table Ch.1.2: Second-Generation Digital Cellular Phone Standards.

Evolution of 2G Systems In 1994 the FCC began auctioning spectrum in the Personal Communication Systems (PCS) band at 1.9 GHz for cellular systems. Operators purchasing spectrum in this band could adopt any standard. Different operators chose different standards. So GSM, IS-136, and IS-95 were all deployed at 1900 MHz in different parts of the country, making nationwide roaming with a single phone difficult. GSM systems operating in the PCS band are sometimes referred to as PCS 1900 systems. Europe allocated additional cellular spectrum in the 1.8 GHz band. The standard for this frequency band called GSM 1800 or DCS 1800 uses GSM as the core standard with some modifications to allow overlays of macrocells and microcells2. Once digital cellular became available, operators began incorporating data services in addition to voice. The 2G systems with added data capabilities are sometimes referred to as 2.5G systems. The enhancements to 2G systems made to support data services are summarized in table Ch.1.3. GSM systems followed several different upgrade paths to provide data services. The simplest called High Speed Circuit Switched Data (HSCSD) allows a maximum transmission rate of up to 57.6 Kbps. Circuit switching is quite inefficient for data, so a more complex enhancement provides for packet-switched data layered on top of the circuit-switched voice. This enhancement is referred to as General Packet Radio Service (GPRS). A maximum data rate of 171.2 Kbps is possible with GPRS when all 8 timeslots of a GSM frame are allocated to a single user. The data rates of GPRS are further enhanced through variable-rate modulation and coding, referred to as Enhanced Data rates for GSM Evolution (EDGE). EDGE provides data rates up to 384 Kbps with a bit rate of 48 to 69.2 Kbps per timeslot. The IS-95 standard was modified to provide data services by assigning multiple orthogonal Walsh functions to a single user. This evolution is referred to as the IS-95b standard. Table Ch.1.3: 2G Enhancements to Support 2.5G Data Capabilities.

Third Generation Systems The fragmentation of standards and frequency bands associated with 2G systems led the International Telecommunications Union (ITU) in the late 1990s to formulate a plan for a single global frequency band

2

Note that second-generation cordless phones such as DECT, the Personal Access Communications System (PACS), and the Personal Handyphone System (PHS) also operate in the 1.9 GHz frequency band, but these systems are mostly within buildings supporting private branch exchange (PBX) services.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

15

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

and standard for third-generation (3G) digital cellular systems. The standard was named the International Mobile Telephone 2000 (IMT-2000) standard with a desired system rollout in the 2000 timeframe. Agreement on a single standard did not materialize, with most countries supporting one of two competing standards: cdma2000 (backward compatible with cdmaOne) supported by the Third Generation Partnership Project 2 (3GPP2) and wideband CDMA (W-CDMA, backward compatible with GSM and IS136) supported by the Third Generation Partnership Project 1 (3GPP1). The main characteristics of these two 3G standards are summarized in table Ch1.4. Both standards use CDMA with power control and RAKE receivers, but the chip rates and other specification details are different. In particular, cdma2000 and W-CDMA are not compatible standards. So a phone must be dualmode to operate with both systems. A third 3G standard, TD-SCDMA, is under consideration in China but is unlikely to be adopted elsewhere. The cdma2000 standard builds on cdmaOne to provide an evolutionary path to 3G. The core of the cdma2000 standard is referred to cdma2000 1X or cdma2000 1XRTT, indicating that the radio transmission technology (RTT) operates in one pair of 1.25 MHz radio channels, and is thus backwards compatible with cdmaOne systems. There are two evolutions of this core technology to provide high data rates (HDR) above 1 Mbps: these evolutions are referred to as cdma2000 1XEV. The first phase of evolution, cdma2000 1XEV-DO (Data Only), enhances the cdmaOne system using a separate 1.25 MHz dedicated high-speed data channel that supports downlink data rates up to 3 Mbps and uplink data rates up to 1.8 Mbps for an averaged combined rate of 2.4 Mbps. The second phase of the evolution, cdma2000 1XEV-DV (Data and Voice), is projected to support up to 4.8 Mbps data rates as well as legacy 1X voice users, 1XRTT data users, and 1XEV-DO data users, all within the same radio channel. Another proposed enhancement to cdma2000 is to aggregate three 1.25 MHz channel into one 3.75 MHz channel. This aggregation is referred to as cdma2000 3X, and its exact specifications are still under development. W-CDMA is the primary competing 3G standard to cdma2000. It has been selected as the 3G successor to GSM, and in this context is refered to as the Universal Mobile Telecommunications System (UMTS). WCDMA is also used in the Japanese FOMA and J-Phone 3G systems. These different systems share the WCDMA link layer protocol (air interface) but have different protocols for other aspects of the system such as routing and speech compression. An enhancement to W-CDMA called High Speed Data Packet Access (HSDPC) provides data rates of around 9 Mbps, and this may be the precursor to 4th-generation systems. The main characteristics of the 3G cellular standards are summarized in table Ch.1.4. Table Ch.1.4: Third-Generation Digital Cellular Phone Standards.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

16

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 02 SIGNAL PROPAGATION AND PATH LOSS MODELS 2.1 RADIO FREQUENCY PROPAGATION OVERVIEW 

Radio-frequency propagation is fuzzy in nature in multipath environments because of irregular terrain, RF barriers, and scattering phenomena (Faruque, S., 1996).



The performance of mobile communication systems is limited by the radio channel and the transmission path between transmitter and receiver and varies randomly from simple line of sight to one obstructed severely by building and foliage (Gibson, D. J., 1996).



Unlike wired channel, it is not easy to analyze the radio channel in wireless medium and is typically done in a statistical fashion based on measurement of any specific communication system (Gibson, D. J., 1996).



The propagation characteristics of electromagnetic (EM) waves in terms of the ability to propagate in a medium vary with environments (Faruque, S., 1996).



Most cellular radio system operates in urban environment where there is no direct wave at the receiver, rather an integrated wave resulting from diffraction, reflection, and scattering from various obstacles (buildings, moving objects, etc.)



Propagation models traditionally focus on the prediction of signal strength at the receiver resulting from various multipath losses (Gibson, D. J., 1996).

Basic Propagation Mechanisms Reflection is a phenomenon (figure Ch2.1) that occurs when an EM wave impinges upon an obstruction with dimensions very large compared to the radio wave wavelength such as earth surface, building, or wall. Diffraction occurs from the signal obstruction obstructed by a surface (earth, buildings, or walls) with sharp irregularities (edges), which may interfere constructively or destructively at the receiver.  According to Huygens principle, all points on a wave front can be considered as point sources for the production of secondary wavelets, and these wavelets combine to produce a new wave front in the direction of propagation (figure Ch 2.2). 

Diffraction is caused by the secondary wavelets into the shadowed region and the strength of the diffracted waves is the sum of all secondary wavelets in the shadowed region.

Scattering occurs when the dimension of the obstacles is small compared to the wavelength and the number of obstacles per unit volume is quite large.

Ref

e c te

av d w

e

R e c e iv e r

T r a n s m itte r

D ire c t w a v e

E a rth G ro u n d D is ta n c e (d )

Figure Ch 2.1: Two ray ground reflection model. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

17

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Q H u y g e n s s e c o n d a ry s o u rc e s

P

K n if ge e ed

S o u rc e

R e c e iv e r in s h a d o w e d re g io n

h

d2

d1

Figure Ch 2.2: Knife-edge diffraction. 2.2 PROPAGATION ENVIRONMENTS Outdoor Propagation Environments 

In an outdoor environment, there are normally classified three types of areas such as urban, suburban, and rural areas, and



two modes such as point to point mode operation (when a detailed terrain path profile, including transmission frequency, path length, polarization, antenna heights, surface refractivity, effective radius of earth, ground conductivity, ground dielectric constant, and climate) and



area to area mode prediction (if the terrain path profile is not available) that provides techniques to estimate the path specific parameters. The terrain profile of a particular area is important and may vary from a simple curved earth to a highly mountainous region.



The presence of trees, buildings, moving cars, and other obstacles and the direct path, reflections from the ground and buildings, and diffraction from the corners and roofs of buildings are the main contributions to the total field generated at a receiver due to radio-wave propagation (Sarkar, K. T., et al., 2003).

Indoor Propagation Environments 

The propagation of signals in indoor environments differ from the outdoor environments mainly in two aspects such as unpredictable interference caused by the electronic equipments and fading effects caused by a hand-held unit. Moreover, the channel characteristics varies (although slowly) because of movements of people and opening and closing of doors.



The propagation characteristics within buildings change strongly by local features such as building’s layout, construction materials, and the type (Gibson, D. J., 1996). The transmitted signal often reaches the receiver through more than one path, due to reflection, refraction, and diffraction of the radio wave by objects inside a building (Sarkar, K. T., et al., 2003). The path loss is given by Q

PL ( d )  PL ( d 0 )  10 n log( d

d0

)



q 1

p

FAF ( q ) 

 WAP

( p)

p 1

WAP ( p ) are the floor and the wall attenuation factors respectively, and d0 where FAF ( q ) and is the close-in distance and d is the separation between transmitter and receiver. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

18

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The propagation path loss (as a function of the distance) also has two distinct regions for indoor environments as shown by the above equation. When the electromagnetic radiation is incident on a wall or a floor in an oblique fashion, less power will be transmitted through the wall than would occur at normal incidence (Sarkar, K. T., et al., 2003). 2.3 OVERVIEW OF PROPAGATION MODELS The path loss is associated with the design of base stations and the signals from the mobile unit received at a base station via multiple paths with different angle of arrival (AOA), path delay, and attenuation in mobile communications. There are two main models (Sarkar, K. T., et al., 2003) for characterizing path loss such as 

Empirical (or statistical) models based on the statistical characterization of the received signal and are easier to implement, require less computational effort, and less sensitive to the environment geometry.



Site-specific (or deterministic) models that have a certain physical basis and in need of a huge amount of data regarding geometry, terrain profile, locations of building and of furniture in buildings, and so on. These models require more computations, and are more accurate.

Currently, a third alternative, which includes many new numerical methods is being introduced to propagation prediction (Sarkar, K. T., et al., 2003). PROPAGATION MODELS AND PATH LOSS ESTIMATION Empirical (or statistical) models Free-Space Propagation Path Loss Model In free-space (no obstacles and atmospheric effects) propagation, the path loss (Gibson, D. J., 1996) is given by the equation,  2 Pr  Pt G t G r (

4 d

)

, where Pr represents received power, Pt represents transmitted power, Gt and Gr represent gain of transmitter and receiver respectively, and d is the distance between transmitter and receiver. But in reality, because of having such obstructions in the way of signal propagation as reflection, diffraction, and scattering of radio waves occurred, which causes multipath propagation (varies with the type and area of obstruction). These signals have longer path than direct signal, and the magnitude as well as the phase difference varies with the path length of the signals. The urban environment is the most common and unpredictable propagation environment in cellular communication systems, which is characterized by dense urban, urban, and suburban environments with the density of civil structure variation. The signal received at the receiver is a result of direct rays, reflected rays, and shadowing as shown in figure Ch 2.3.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

19

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Transmitter Multipath signal

Dire

ct pa

th Sig nal

Receiver

Figure Ch 2.3 A typical urban environment showing direct path and multipath propagation. Shadowing is defined as the result of signal obstruction by trees, foliage, etc. (figure Ch 2.4), which is difficult to exact measure as the density of trees, foliage, etc. is variable in place and season. In addition, the magnitude of the loss due to shadowing depends on the density of forest and seasonal variations. 20

Figure Ch 2.4 A mobile environments showing shadowing effect (Faruque, S. (1996). Log-Distance Path Loss Model The average received power decreases with distance raised to some extent (based on both theoretical and measurement based propagation models). The average path loss for an arbitrary distance between transmitter and receiver is expressed (in this model) as a function of distance (Gibson, D. J., 1996) using a path loss exponent. n

 d   PL ( d )    d   o  ,

and in decibels, PL  PL  d o   10 n log

10

 d    d   o ,

where n represents path loss exponent (the rate at which the path loss increases with distance), Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

d0 is the free space reference distance (for macro cell, d0 > 1km, and micro cell d0 < 1 km), and d is the distance between transmitter and receiver. The path loss is a straight line with a slope n when plotted on a log-log scale, and the value of n depends on the degree of obstruction in the wave the signal propagates. The table Ch 2.1 below gives some typical path loss exponents in various mobile radio environments. Table Ch 2.1 Path loss exponents for different environments (Gibson, D. J., 1996). Environment Free space Urban area cellular radio Shadowed urban cellular radio Line of sight in building Obstructed in building Obstructed in factories

Path Loss exponent n 2 2.7 - 4 5-6 1.6 – 1.8 4-6 2-3

Based on the measurements, at any value of d, the path loss PL (d) for any particular location is lognormally distributed about the mean distance dependent value and is given by PL  PL ( d )  X  ,

where X  represents zero mean log normally distributed (showing shadowing effects) random variable with standard deviation  . Hence, a path loss exponent n and standard deviation  statistically describes the path loss for any arbitrary location having specific distance d between the transmitter and the receiver. Okumura Model Okumura’s model is widely used models for signal prediction in urban areas, which is applicable in the frequency range of 150 MHz – 2 GHz, the distance of 1 – 100 km, and the base station effective antenna height of 30 m – 1000 m over a quasi smooth terrain. Okumura developed a set of curves representing median attenuation relative to free space (A mu) in an urban area (Figure Ch 2.3) with base station height (hte) of 200 m and a mobile station height (hre) of 3 m, which are plotted as a function of frequency (100 MHz - 3000 MHz) and distance (1 km - 100 km from the base station). The model is given by (Gibson, D. J., 1996) L50 = LF + Amu (f, d) + G(hte) + G(hre), where L50 represents median propagation loss in dB, LF represents free space loss, G(hte) represents base station antenna height gain factor, and G(hre) represents mobile antenna height gain factor.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

21

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure Ch 2.5 Median attenuation relative to free-space over a quasi smooth terrain developed by Okumura. 22

Various corrections can be made to the model based terrain related parameters such as base station antenna height, terrain undulation height, isolated ridge height, terrain average slope, and mixed land-sea parameters. And the necessary correction factors (Figure Ch 2.4) can be added or subtracted after the calculation of parameters. Features (of Okumura model)  Practically, graphical path loss calculation.  Urban environment consideration.  Measured-data basis. Advantages (of Okumura model)  Simplest and best in terms of accuracy in predicting path loss for early cellular systems (Sarkar, K. T., et al., 2003).  Very practical model used now a day as standard for mobile radio systems planning in Japan (Sarkar, K. T., et al., 2003). Limitations (of Okumura model)  The model does not provide any analytical explanation and is based on only measured data.  The model is complex and provides slow response to rapid changes in radio path profile.  Well applicable in urban and suburban areas and not as good in rural areas (over irregular terrain).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure Ch 2.6 Correction factor for different type of terrain developed by Okumura. 23

Hata Model The Hata model (Hata, 1980) is the empirical formulization of the graphical path loss information provided by Okumora. This model provides a standard formula for path loss in urban environment and correction equations for other environments (suburban and rural) and is given by (Gibson, D. J., 1996) L 50 ( urban )  69 . 55  26 . 16 log

10

f c  13 . 82 log

10

h te  a ( h re )  ( 44 . 9  6 . 55 log

10

h te ) log

10

d

where L50 (path loss) in dB, fc (frequency), 150 MHz-1500 MHz, hte (effective base station height), 30m-200 m, hre (mobile antenna height), 1m-10m, d (distance between transmitter and receiver, 1 – 20 km), and a(hre) is the correction factor for effective mobile antenna height (function of the service area or city). For small to medium sized city, a ( h re )  (1 . 1 log

10

f c  0 . 7 ) h re  (1 . 56 log

10

f c  0 . 8 ) dB

And for a large city a ( h re )  {8 . 29 (log a ( h re )  [ 3 . 2 {log

10

10

1 . 54 h re )

2

(11 . 75 h re )}

 1 . 1} dB  ( f c  300 MHz ) 2

 4 . 97 ] dB  ( f c  300 MHz )

For suburban area, the path loss is give by L 50  L 50 ( urban )  2 [log

10

(

fc

)]  5 . 4 dB 2

28

For open areas (rural) the formula is modified as L 50  L 50 ( urban )  4 . 78 (log

f c )  18 . 33 log 2

10

10

f c  40 . 98

dB

The open area of the Hata model corresponds to flat deserted area. For path loss of typical rural area a margin of 6 - 10 dB is often added to the path loss predicted by the open area Hata model. In rural areas Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

with heavy forest also a 10 dB attenuation should be added to the path loss to compensate for (summertime) foliage loss. In Hata model a large city is understood to be heavily built with relatively large buildings averaging more than four floors in height. If the city has lower average building height, it is considered to be small or medium. For higher carrier frequencies of 1500 – 2000 MHz, the following modification of Hata model for urban area has been proposed. L 50 ( urban )  46 . 3  33 . 9 log 10 f c  13 . 82 log 10 h te  a ( h re )  ( 44 . 9  6 . 55 log 10 h te ) log 10 d  C M

where additional correction factor is described by  0 dB CM    3 dB

for medium

sized cities and suburban

areas

.

for metropolit an centers

These modified equations have been successfully used for cellular mobile network design at 1800 MHz band. However, it should be noted that (modified) Hata model is only valid for macrocell design. Hata model is not applicable to microcells with d < 1 km. The Hata model has also been extended to distances d = 20 - 100 km with the following modification. L 50 ( urban )  69 . 55  26 . 16 log 10 f c  13 . 82 log 10 h te  a ( h re )  ( 44 . 9  6 . 55 log 10 h te ) log 10 d 

where

 

α  1  0 .14  0 . 000187 f  0 . 00107 h te  log d

20



0 .8 . 24

Hata model is an example of an area-to-area path loss prediction model, which is used to predict a path loss over a generalized terrain without knowing the particular terrain configuration between BS and MS. Area-to-area models can usually provide only an accuracy of prediction within a standard deviation of 8 dB. The 8 dB spread of path loss remains constant irrespective of distance. It is found from log-normal distribution of the path loss that 68 % of the values fall within a range of ±8 dB. The uncertainty range of area-to-area models is too large for network design Point-to-point prediction models can reduce the uncertainty range by including the detailed terrain contour information to the path loss prediction. Especially for microcells (radius less than 1 km) the contribution of individual buildings for path loss is very important. Walfisch and Bertoni Model This model takes into consideration of the rooftops and the height of the buildings to predict the received signal level (strength) (Gibson, D. J., 1996) at the street using diffraction phenomenon (see figure 2.6). The path loss is given by L p  P0 Q P1 2

, where P0 is the free space path loss (ratio of the received power to the radiated power for isotropic antennas in free space) and is given by 2

   P0    .  4 d  2

represents the reduction in the rooftop signal at the row of buildings that shadows the receiver at the street level, and P1 represents the signal loss from the rooftop to the street based on diffraction. Q

In dB the path loss equation can be written as Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

L p [ dB ]  L 0 [ dB ]  L rts [ dB ]  L ms [ dB ]

where L 0 represents free space loss, L rts represents rooftop- to-street diffraction and scatter loss, and L ms represents multi screen diffraction loss due to the building rows.

Figure Ch 2.7 Propagation geometry for model proposed by Walfisch and Bertoni.

25

Features (of Walfisch and Bertoni model)  The model considers the impact of rooftops and building height using diffraction.  Simple formulation taking into account of only three factors. Limitations (of Walfisch and Bertoni model)  The model does not formalize any specific area based path loss prediction, rather more generalized prediction of path loss.  The model is based on only diffraction phenomenon (reflection, terrain profile, foliage, and other impacts do not specify).  It does not mention about the distance (d) in between transmitter and receiver and applicability about cell coverage area (macrocell, microcell, picocell, etc). Advantages (of Walfisch and Bertoni model)  Simple formulation of path loss prediction.  Suitable for urban areas (because of high density both in buildings, movable device, etc.). COST-231-Walfisch-lkegami Model This model is based on Walfisch-Bertoni model consisted of three terms and is given by L b  ( L 0  L rts  L msd )

for L rts  L msd  0

Lb  L0 and for L rts  L msd ≤ 0 where L0 represents the free space loss, Lrts represents roof-top-to-street diffraction and scattering loss, and Lmsd represents multi-screen diffraction loss.

The free space loss is given by Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

L 0  32 . 4  20 log d  20 log f

, where d is the distance (in km) between transmitter and receiver and f is the radio frequency (in MHz). The roof-top-to-street diffraction and scattering loss L rts   16 . 9  10 log w  10 log f  20 log  h Roof  L ori

,

where w is the street width (in m) and  h Roof  h Roof  h Mobile . hRoof is the height of the building and hm (or hMobile ) is the height of the mobile antenna (see figure Ch 2.8).

26

Figure Ch 2.8 Walfisch and Ikegami Model parameters representation (Ahmed, M. K., 2010). The orientation loss, Lori is given by  10  0 . 354  L ori  2 . 5  0 . 075 (  35 )

0    35 0

0

0 0 for 35    55

4 . 0  0 . 114 (  55 )

55    90 0

0

where  is the angle of incidence relative to the direction of the street (See Figure Ch 2.8).

Figure Ch 2.9 Angle of incidence  in Walfisch and Ikegami Model. The multi screen diffraction loss, L msd is given by L msd  L bsh  k a  k d log d  k

f

log f  9 log b

where b is the distance between the buildings along the signal path L msd and k a represent the increase in path loss due to reduced base station antenna height and are given by

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

h Base  h Roof

54 k a  54  0 . 8  h Base

d  0 . 5 km and h Base  h Roof

54  1 . 6  h Base d d  0 . 5 km and h Base  h Roof

L bsh 

 18 log( 1   h Base )

h Base  h Roof

0

h Base  h Roof

where hb or (hBase ) is the base station height and Lbsh is the shadowing gain. kd and kf control the dependence of the multi-screen diffraction loss as a function of distance and the radio frequency of operation respectively. So are given by for medium-sized cities and suburban centers with moderate tree densities. h Base  h Roof

18 kd 

18  15

 h Base h Roof

and

h Base  h Roof

k f   4  0 .7 (

f

 1)

925

But for metropolitan centers, kf is given by k f   4  1 .5 (

f

 1)

27

925

This model is being considered for use by ITU-R in the International Mobile Telecommunications - 2000 (IMT- 2000) standards activities (Sarkar, K. T., et al., 2003). Limitations  The model is suitable when h b 

 h Roof

.

For equal and less heights, large prediction error can be expected.

Site-Specific Models for Path Loss Ray-Tracing Technique Ray tracing is a technique based on Geometrical Optics (GO) that can be applied as an approximate method for estimating the levels of high-frequency electromagnetic fields. In GO, it assumes that energy can be considered to be radiated through infinitesimally small tubes, often called rays, and they lie along the direction of propagation and travel in straight lines (when the relative refractive index of the medium is constant). Therefore, signal propagation can be modeled via ray propagation (Sarkar, K. T., et al., 2003). This model can be used with sufficient precision to predict radio coverage for large buildings having a large number of walls between the transmitter and the receiver. A ray-tracing algorithm provides a relatively simple solution for radio propagation based on GO. However, Ray-tracing fails to correctly predict the scattered fields for complex lossy structures with finite dimensions. In the following, the 2-ray model Ground Reflection is described. 2-ray Ground Reflection Model The 2-ray ground reflection model, shown in figure Ch 2.9, is a useful propagation model that considers both the direct path and a ground reflected propagation path between transmitter and receiver. This model has been found to be reasonably accurate for predicting the large-scale signal strength over distances of

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

several kilometers for mobile radio systems that use tall towers (heights which exceed 50 m) as well as for line-of-sight microcell channels in urban environments.

Figure Ch 2.10 Two-ray ground reflection model. The maximum T- R separation distance normally is at most only a few tens of kilometers, and it is assumed that the earth is flat. The total received E-field, ETOT, is then a result of the direct line-of-sight component, ELOS and the ground reflected component, Eg. Consider Pt and Pr are respectively the transmit power and the receive power, ht is the height of the transmitter, hr is the height of the receiver, and Gt and Gr are respectively the gain of transmit antenna and receive antenna. The received power at a distance d from the transmitter can be expressed as3 2

Pr  Pt G t G r

28

2

ht hr d

4

Note that this final equation for the average received power from the two-ray model has the desired 1 d

4

dependency on distance. 2.4 SUMMARY

A brief overview of the various propagation models dealing with path loss has been given. Propagation models dealing with path loss have been emphasized using two approaches; empirical or statistical model of the path loss where some of the parameters used were determined empirically from measurements, and sitespecific methods where ray tracing is the main method with some other numerical methods used in electromagnetic-field computation have also been explained. Each of these approaches makes a very different trade-off of accuracy versus complexity (Sarkar, K. T., et al., 2003). The empirical (statistical) models are extremely simple (no environmental information is used other than in the choice of the parameters), but lacks from accuracy; whereas site-specific models are considerably more accurate but they require a great deal of specific information about the area of interest.

3

The derivation of Pr can be found in Wireless communications, Theodore S. Rappaport, 2002, Pages 98 – 101.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 03 CONCEPT OF FADING IN CELLULAR RADIO SYSTEMS 3.1 MULTIPATH FADING For most practical channels, where signal propagation takes place in the atmosphere and near the ground, the free space propagation model is inadequate to describe the channel and predict system performance. The received signals from various radio paths are summed at the MS antenna. This summing can be either constructive or destructive. In the destructive case, the sum signal has power that is clearly smaller than the average level of the received signal, and the received signal is said to be in a fade. The effect can cause fluctuations in the received signal’s amplitude, phase, and angle of arrival. The overall phenomenon of radio propagation is called multipath fading. Figure 3.1 represents an overview of fading channel manifestations. It starts with two types of fading effects that characterize mobile communications: large-scale and small-scale fading.

29

Figure 3.1 Different types of fading. Large-scale fading Large-scale fading represents the average signal power attenuation or path loss due to motion over large areas. This phenomenon is affected by prominent terrain contours (hills, forests, billboards, clumps of buildings, etc.) between the transmitter and receiver. The receiver is often represented as being “shadowed” by such prominences. The statistics of large-scale fading provide a way of computing an estimate of path loss as a function of distance. This is described in terms of a mean-path loss (nth-power law) and a lognormally distributed variation about the mean. Mathematically, L p ( d )( dB )  L s ( d 0 )  10 n log 10 ( d )  X  ( dB ) d0

where is the mean path loss as a function of distance d between the transmitter and receiver. The path loss Ls (d0) to the reference point at a distance do from the transmitter is typically found through field measurements or calculated using the free-space path loss. L p (d )

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

d0 is the reference distance. The reference distance do corresponds to a point located in the far field of the antenna. Typically, the value of do is taken to be 1 km for large cells, 100 m for microcells, and 1 m for indoor channels. X  denotes a zero-mean Gaussian random variable (in decibels) with standard deviation  (also in decibels) is site- and distance-dependent and is often based on measurements. It is not unusual for it to take on values as high as 6-10 dB or greater. Small-scale fading Small-scale fading refers to the dramatic changes in signal amplitude and phase that can be experienced as a result of small changes (as small as a half-wavelength) in the spatial separation between a receiver and transmitter. Small-scale fading manifests itself in two mechanisms, namely, time-spreading of the signal (or signal dispersion) and time-variant behavior of the channel. For mobile radio applications, the channel is time-variant because of motion between the transmitter and receiver that results in propagation path changes. The rate of change of these propagation conditions accounts for the fading rapidity (rate of change of the fading impairments). Small-scale fading is also called Rayleigh fading because if the multiple reflective paths are large in number, and there is no line-of-sight signal component, the envelope of the received signal is statistically described by a Rayleigh distribution as follows. Figure 3.2 depicts the Rayleigh pdf. p (r ) 

r

r



2

e

2

2

2

,

for r  0

p (r )  0, for r< 0 2 where r is the envelope amplitude of the received signal, and ς is the predetection mean power of the multipath signal. For a single link it represents the pdf associated with the worst case of fading per mean received signal power (figure 3.3). However, when there is a dominant non fading signal component present such as a line-of-sight propagation path, the small-scale fading envelope is described by a Rician pdf.

Figure 3.2 Raleigh distribution.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

30

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 3.3 A typical Rayleigh fading envelope at 900 MHz, 120 km/hr. A mobile radio roaming over a large area must process signals that experience both types of fading: small-scale fading superimposed on large-scale fading. Figure 3.4 illustrates the various contributions that must be considered when estimating path loss for a link budget analysis in a cellular application. These contributions are:  Mean path loss as a function of distance due to large- scale fading,  Near-worst-case variations about the mean path loss (typically 6-10 dB), or large-scale fading margin, and  Near-worst-case Rayleigh or small-scale fading margin (typically 20-30 dB). Note that in figure 3.4, the annotations = 1-2% indicates a suggested area (probability) under the tail of each pdf as a design goal. Hence, the amount of margin indicated is intended to provide adequate received signal power for approximately 98-99 percent of each type of fading variation (large- and smallscale).

Figure 3.4 Link budget considerations for a fading channel. A received signal r(t) is generally described in terms of a transmitted signal s(t) convolved with the impulse response of the channel hc(t). Neglecting the degradation due to noise, we write r ( t )  s( t )  h c ( t )

, where * denotes convolution. In the case of mobile radios, r (t) can be partitioned in terms of two component random variables, as follows.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

31

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

r ( t )  m ( t )  r0 ( t )

, where m(t) is called the large-scale-fading component and ro(t) is called the small-scale-fading component. m(t) is sometimes referred to as the local mean or log-normal fading because the magnitude of m(t) is described by a log-normal pdf (or equivalently the magnitude measured in decibels has a Gaussian pdf). r0 is sometimes referred to as multipath or Rayleigh fading. Figure 3.5 illustrates the relationship between large-scale and small-scale fading. In Figure 3.5 (a), received signal power r(t) versus antenna displacement (typically in units of wavelength) is plotted for the case of a mobile radio. Small-scale fading superimposed on large-scale fading can be readily identified. The typical antenna displacement between the small-scale signal nulls is approximately a half wavelength. Note that, In Figure 3.5 (b) the large scale fading or local mean m(t) has been removed in order to view the smalls scale fading r0(t) about some average constant power.

32

Figure 3.5 large-scale fading and small-scale fading. Fast fading Fast fading occurs in all situations where the receiver is located among buildings, vehicles and other objects that reflect or scatter radio waves. Fast fading will even be observed by a stationary receiver operating on a fixed frequency due to motion of scatterers such as people, vehicles, and trees in the vicinity of the receiver. Fast fading has a scale length of one half-wave length (fades are on average separated by about one halfwavelength), and results from the interference when signals arrive at the receiver from many directions. When the fast fading is smoothed by averaging over spatial length of 20 to 40 wavelengths (about 5 –10 m at 1 –2 GHz carrier frequencies), the result, which is called the sector average (figure 3.6), shows a variation over a scale length on the order of 10 m (order of building widths). When the measured sector averages are plotted in dB versus distance from the base station on a logarithmic scale, the deviation of the sector average from the mean value (that corresponds to path loss) is typically found to have a Gaussian distribution in dB scale (figure 3.7).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 3.6 Concept of sector average.

33

Figure 3.7 Deviation of the sector average from the mean value. Two very important parameters of fast fading are  the level crossing rate and  the average duration of fades. The knowledge of the average fade duration is important for choosing appropriate error control coding methods for digital transmission. A fade is encountered when the level of the received signal is below some specified limit r. For the level crossing rate, we calculate the positive slopes at level r (figure 3.8). The total number of crossings within a time interval divided by the length of the time interval becomes the level crossing rate (lcr).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 3.8 level crossing rate and average fade duration. In Rayleigh fading environment, we can find the level crossing rate n(r) from the following equation. n r  

N



T

 fm

r



r

2 2

e

34

2

, where fm = v/λ is the maximum Doppler frequency, and ς2 is the variance of the in-phase and quadrature components of the Rayleigh fading signal. lcr is the expected rate at which the Rayleigh fading envelope normalized to the local rms signal level crosses a specified level in a positive-going direction. lcr depends on upon the rate of motion. This rate of motion is normally characterized by the Doppler spread: the maximum frequency shift in the received signal due to relative motion of the transmitter, receiver, and objects in the channel. The average duration of fades (adf) is the sum of durations of individual fades divided by their number.

t r  

T

N  ti i 1

N

T



1



 fm r

r

[e

2 2

2

 1]

, which includes an important result n(r). t(r) = F(r), where F(r) is the Rayleigh cdf. If we normalize the level crossing rate n(r) by t0 

1

n0 

2

v



and the average fade duration t(r) by

, we can plot them as a function of the signal level with respect to the rms level,

n0

r 2

,independent

of the carrier frequency and the vehicle speed (figure 3.9).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

35

Figure 3.9 level crossing rate n(r) and the average fade duration t(r) a function of the signal level with respect to the rms level

r 2

.

Slow fading The variation of average level is referred to as slow fading (or shadow loss). Slow fading (or shadow loss) describes the random shadowing effect among a large number of locations, which have the same distance between MS and BS but have different mixtures of objects in the propagation path.

 d  vt

Figure 3.10 Slow and fast fading. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

The slow fading (or shadow loss) is due to variations in the shadowing of the receiver as it moves around in the local urban environment, e.g. past buildings of different heights, from ground level to bridge, or elevated highway. Treated as a random variable, the variation can have a standard deviation as high as 8 dB. Viewed in other way, the slow fading is the error inherent in the simple path loss calculation model. In radio propagation from BS to MS, several multiplicative random processes act on the signal. Thus in dB scale, the random gains are summed. The sum of several random variables tends towards Gaussian distribution in dB scale that corresponds to lognormal distribution of the received level in the linear scale.

36

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 04 CHANNEL CLASSIFICATION IN CELLULAR RADIO SYSTEM 4.1 ILLUSTRATION OF FADING

37

Figure 4.1 Different types of fading. 4.2 CLASSIFICATION OF CHANNELS A mobile radio channel generally exhibits both Delay and Doppler shift spreading. Delay spreading corresponds to time dispersion in time domain and to frequency-selective fading in frequency domain. Doppler spreading corresponds to frequency dispersion in frequency domain and to time-selective fading in time domain. These effects are shown by any mobile radio channel, but their importance varies according to the symbol duration and the bandwidth of the transmitted signal. Fading Channel Characteristics Power-Delay Profile Random and complicated radio-propagation channels can be characterized using the impulse-response approach. For each point in the three-dimensional environment, the channel is a linear filter with impulse response h(t). The impulse response provides a wideband characterization of the propagating channel, and contains all of the information necessary to simulate or analyze any type of radio transmission through that channel. If the input signal is a unit impulse  (t ) , the output will be the channel impulse response, which can be written as

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

where An,  n and  n are the attenuation, delay in time of arrival, and phase, corresponding to path n respectively. Multipath propagation causes severe dispersion of the transmitted signal and the expected degree of dispersion is determined through the measurement of the power-delay profile of the channel. The power-delay profile provides an indication of the dispersion or distribution of transmitted power over various paths in a multipath model for propagation. The power-delay profile of the channel is calculated by taking the spatial average of

h t 

2

over a

local area. By making several local-area measurements of h t  for different locations, it is possible to build an ensemble of power delay profiles, each one representing a possible small-scale multipath channel state. A typical plot of the power-delay profile is shown in figure 4.2. Many multipath-channel parameters are derived from the power-delay profile. A mobile channel exhibits a continuous multipath structure. Hence, the power-delay profile can be thought of as a density function of the form 2

38

Delay Spread The delay spread δ of a channel is the delay interval where most of the power in the power delay profile4 Ph (τ) of the channel is concentrated.

where  is the mean delay. And is given by

4

We use P h   and P   interchangeably to represent power delay profile.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Typical values for delay spread in different environments are given in table 4.1. Table 4.1 Typical delay spread in different environments.

The parameters which characterize the delay spread can be classified as  First-Arrival delay,  Mean access delay,  RMS delay spread, and  Excess delay spread.

39

(a)

(b) Figure 4.2 An illustration of a typical power-delay profile and the definition of the delay parameters.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

First-Arrival Delay (  A ) This is a time delay corresponding to the arrival of the first transmitted signal at the receiver. It is usually measured at the receiver. This delay is set by the minimum possible propagation path delay from the transmitter to the receiver. It serves as a reference, and all delay measurements are made relative to it. Any measured delay longer than this reference delay is called an excess delay. Mean Excess Delay (  e or  ) This is the first moment of the power-delay profile (as shown in figure 4.2) with respect to the first delay. It is expressed as

RMS Delay (  RMS or   )

40

This is the square root of the second central moment of a power-delay profile, as seen in Figure4.2. It is the standard deviation about the mean excess delay, and is expressed as

The RMS delay is a good measure of the multipath spread. It gives an indication of the nature of the inter-symbol interference (ISI). It is also used to give an estimate of the maximum data rate for transmission. Maximum Excess Delay (  m ) This is measured with respect to a specific power level, which is characterized as the threshold of the signal. When the signal level is lower than the threshold, it is processed as noise. For example, the maximum excess delay spread can be specified as the excess delay (  m ) for which P (  ) falls below -30 dB with respect to its peak value, as shown in Figure 4.2. Thus, maximum excess delay  m is Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

where  0 is the first arriving signal and  X is the maximum delay at which a multipath component is within X dB of the strongest arriving multipath signal (which does not necessarily arrive at  0 ). Time Dispersion and Frequency-Selective Fading Time dispersion and frequency-selective fading are both manifestations of multipath propagation with delay spread. Presence of one implies the presence of the other. Time dispersion stretches a signal in time so that the duration of the received signal is greater than that of the transmitted signal. Frequency selective fading filters the transmitted signal, attenuating certain frequencies more than the others. Two frequency components closely spaced receive approximately the same attenuation. However, if they are far apart, they often receive vastly different attenuations. In digital systems it results as inter symbol interference. The minimum transmission bandwidth at which time dispersion is observable is inversely proportional to the maximum excess delay of the channel τm, where the excess delay is the actual delay minus the delay of the first arrival path. The constant of proportionality is normally taken as ¼, although it is system dependent. So, delay spread has 2 observable effects: distortion and dispersion. 4.3 COHERENCE BANDWIDTH Coherence Bandwidth (Bc) is the measure of maximum possible transmission bandwidth at which distortion becomes appreciable. Bc indicates the frequency separation at which the attenuation of amplitudes of two frequency components become decorreleted such that the envelope correlation coefficient ρ (Δf, Δt) reaches a pre designated value. This value can vary from 0.9 to 0.37. Note that a value of 0.5 can safely be taken for mobile communications. We can define

, where ‹ › denotes the ensemble average. a1 and a2 represent the amplitudes of signals at frequencies f1 and f2 respectively and at time t1 and t2 respectively. Note that|f1 - f2|= Δf and |t1- t2| = Δt. Coherence bandwidth can be characterized as  Range of frequencies over which the channel can be considered flat, meaning that the channel passes all spectral components with approximately equal gain and linear phase.  Frequency components in this bandwidth have a strong correlation in amplitude. For envelopes of two signals to vary uncorrelatedly their frequencies should be separated by more than the coherence bandwidth of the channel Bc. The coherence bandwidth can be approximately calculated from the envelope correlation coefficient between two signals separated by Δf Hz and Δt seconds. When we consider correlation as function of frequency separation only and set Δt to zero, the coherence bandwidth Bc is defined as the bandwidth Δf, where the envelope correlation coefficient between two signals has fallen to one half of its maximum value.

The solution to this is

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

41

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

. A typical delay spread value of 3 μs in urban environment corresponds to coherence bandwidth of about 50 kHz. The typical coherence bandwidths in different kinds of manmade environments are:

Frequency-flat and Frequency-selective fading If the bandwidth of the modulated signal is less than the coherence bandwidth of the channel, all the frequency components of the signal encounter (approximately) same fading and the fading is called frequency-flat fading. On the other hand, if the bandwidth of the modulated signal is much greater than the coherence bandwidth of the channel, different frequency components of the signal encounter different fading characteristics and the fading is called frequency-selective fading. Frequency-selective channels are also called time dispersive channels, because the long delay spread corresponds to lengthening of the duration of the transmitted symbols. In this case the channel has clear filtering effect on the transmitted pulse. And besides the amplitude, also the shape of the pulse is changed5. In digital mobile communications, the propagation phenomena are highly dependent on the ratio of the symbol duration to the delay spread of the time variant radio channel. If the transmission bit rate is so high that each data symbol significantly spreads into adjacent symbols, severe inter symbol interference (ISI) occurs. If we want the interference between adjacent symbols to be low, we have exactly the same equation for maximum transmission symbol rate as for coherence bandwidth, i.e.

. Thus, for typical urban delay spread of 3 μs, the maximum transmission symbol rate without significant (ISI) for binary symbols are 53 kbit/s. For higher symbol rates the ISI must be decreased by equalizers to provide acceptable BER. It should be noted that if we consider the phase correlation between two signals instead of the envelope correlation, the coherence bandwidth is different as follows from the one given immediate before this.

4.4 DOPPLER SPREAD Doppler Spread (BD) or Spectral broadening is caused by the relative movement of the mobile and the base station. Note that power density spectrum of a sine wave suffering from a Doppler spread. Figure 4.3 shows a Doppler power spectral density S(v) plotted as a function of Doppler-frequency shift v.

5

Note that shadowing (slow fading) is always frequency-flat, whereas fast fading due to multiple paths typically causes frequency-selective fading. Thus, the effect of shadowing is independent of the signal bandwidth, but the effect of fast fading depends on it. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

42

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 4.3 Doppler power spectrum. For the case of the dense-scatterer model, a vertical receive antenna with constant azimuthal gain, a uniform distribution of signals arriving at all arrival angles throughout the range (0, 2  ), and an unmodulated CW signal, the signal spectrum at the antenna terminals is given by

. The equality holds for frequency shifts of v that are in the range  fd about the carrier frequency fc and would be zero outside that range. The shape of the RF Doppler spectrum described by the above equation is classically bowl-shaped, as seen in figure 4.3. Note that the spectral shape is a result of the densescatterer channel model. And the dense-scatterer model does not hold for the indoor radio channel; the channel model for an indoor area assumes S (v) to be a flat spectrum. In figure 4.3, the sharpness of the boundaries of the Doppler spectrum are due to the sharp upper limit on the Doppler shift produced by a vehicular antenna traveling among the stationary scatterers of the dense scatterer model. The largest magnitude (infinite) of S(v) occurs when the scatterer is directly ahead of the moving antenna platform or directly behind it. In that case, the magnitude of the frequency shift is given by

, where V is relative velocity, and  is the signal wavelength. fd is positive when the transmitter and receiver move toward each other, and negative when moving away from each other. Knowledge of S(v) allows us to gather how much spectral broadening is imposed on the signal as a function of the rate of change in the channel state. The width of the Doppler power spectrum is referred to as the spectral broadening or Doppler spread denoted by fd and sometimes called the fading bandwidth of the channel6. 4.5 COHERENCE TIME The Doppler spread fd and the coherence time TO are reciprocally related (within a multiplicative constant. When T0 is defined more precisely as the time duration over which the channel's response to a sinusoid has a correlation greater than 0.5, the relationship between T0 and fd is approximately

6

In a typical multipath environment, the received signal arrives from several reflected paths with different path distances and different angles of arrival, and the Doppler shift of each arriving path is generally different from that of another path. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

43

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

. An implication of coherence time to mobile system design is that the radio channel changes its characteristics within a time interval that is of order of the coherence time. Thus, adaptive parts of the receiver (e.g. adaptive equalizer) must be able to adapt within this time frame. Time-flat and time-selective Fading If the symbol duration in the modulated signal is less than the coherence time of the channel, the channel appears to a single symbol to be time-invariant, and the fading is called time-flat fading. On the other hand, if the symbol duration of the modulated signal is much greater than the coherence time of the channel, different time segments of the symbol encounter different fading characteristics, and the fading is called time-selective fading. Time-selective channels are also called frequency dispersive channels, because the large Doppler shift corresponds to broadening of the bandwidth of the transmitted signal. 4.6 CHANNEL CLASSES Now we can define the channel class based on the parameters described. The coherence bandwidth Bc and coherence time Tc are the parameters of the channel that define the effect of channel to different transmitted signals. If the bandwidth of the transmitted signal is less than the coherence bandwidth, the channel is viewed by the system as having a flat frequency response over the transmission bandwidth and is therefore referred to as being frequency-flat. Similarly, if the duration of the transmitted symbol is less than the coherence time, the channel is viewed by the system as being time-invariant and is therefore referred to as being time-flat. The figure 4.4 (a, b) shows the classification of channels following this approach. The curve in the figure illustrates the basic uncertainty principle that the duration-bandwidth product of any signal cannot be less than 1/4π. A more rigorous system of classification, emphasizing the differences between distorting and dispersive channels is shown in figure 4.4 (c).

(a)

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

44

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

(b)

45

(c) Figure 4.4 Channel classification. From this classification it is obvious that wideband communication systems are normally affected by frequency-selective fading (figure 4.5), which appears in high-rate digital systems as inter symbol interference. Narrowband systems encounter frequency-flat characteristics (figure 4.6) and mainly suffer from timeselective fading.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The situation where the channel is flat with neither time nor frequency does not occur with mobile channels because of the short delay spreads and low mobile speeds7.

Figure 4.5 Frequency selective fading channel characteristics. 46

Figure 4.6 Frequency flat fading channel characteristics.

7

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 05 CELLULAR SYSTEM DESIGN 5.1 CELLULAR SYSTEM DESIGN BACKGROUND AND CONCEPT The system design objective of early mobile radio systems was to achieve a large coverage area by using a single, high powered transmitter with an antenna mounted on a tall tower. However, with this approach, it was impossible to reuse the same frequencies throughout the system because of frequency reuse results in interference. A good example is the Bell mobile system in New York City (in the 1970s) that could only support a maximum of twelve simultaneous calls over a thousand square miles. In addition, government regulatory agencies could not make spectrum allocations in proportion to the increasing demand for mobile services. Considering these constraints, it became essential to restructure the radio telephone system to achieve high capacity with limited radio spectrum that covers very large areas as well. And the cellular concept was applied in restructuring the early radio telephone system. The cellular concept is a system level idea where  A single, high power transmitter (large cell) is replaced with many low power transmitters (small cells) and each small cell provides coverage to only a small portion of the service area. 

Each base station is allocated a portion of the total number of channels available to the entire system.



And nearby base stations is assigned different groups of channels so that all the available channels are assigned to a relatively small number of neighboring base stations, and the interference between base stations (and the mobile users under their control) is minimized.



Base stations and their channel groups are systematically spaced throughout a market so that the available channels are distributed throughout the geographic region and may be reused as many times as necessary so long as the interference between co-channel stations is kept below acceptable levels.



With the demand for service increases, i.e. as more channels are needed within a particular market), the number of base stations may be increased along with a corresponding decrease in transmitter power to avoid added interference.



Thereby, it provides additional radio capacity with no additional increase in radio spectrum – the fundamental principle, which is the foundation for all modem wireless communication systems. Hence, the cellular concept enables a fixed number of channels to serve an arbitrarily large number of subscribers by reusing the channels throughout the coverage region. Furthermore, the cellular concept allows every piece of subscriber equipment within a country or continent to be manufactured with the same set of channels, so that any mobile may be used anywhere within the region. Cellular Frequency Reuse Concept Conceptual Model 

In cellular system, each cellular base station is allocated a group of radio channels to be used within a small geographic area called a cell.



Base stations in adjacent cells are assigned channel groups which contain completely different channels than neighboring cells. The base station antennas are designed to achieve the desired coverage within the particular cell.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

47

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



By limiting the coverage area to within the boundaries of a cell, the same group of channels may be used to cover different cells that are separated from one another by distances large enough to keep interference levels within tolerable limits.



The design process of selecting and allocating channel groups for all of the cellular base stations within a system is called frequency reuse or frequency planning. Figure 5.1 illustrates the concept of cellular frequency reuse, where cells labeled with the same letter use the same group of channels.

Note that the hexagonal cell shape shown in figure 5.1 is conceptual and is a simplistic model of the radio coverage for each base station. However, it has been universally adopted since the hexagon permits easy and manageable analysis of a cellular system. The actual radio coverage of a cell is known as the footprint and is determined from field measurements or propagation prediction models. Although the real footprint is amorphous in nature, a regular cell shape is needed for systematic system design and adaptation for future growth. While it might seem natural to choose a circle to represent the coverage area of a base station, adjacent circles cannot be overlaid upon a map without leaving gaps or creating overlapping regions. Thus, when considering geometric shapes which cover an entire region without overlap and with equal area, there are three sensible choices: a square; an equilateral triangle; and a hexagon. However, for a given distance between the center of a polygon and its farthest perimeter points, the hexagon has the largest area of the three. Thus, by using the hexagon geometric, the fewest number of cells can cover a geographic region, and the hexagon closely approximates a circular radiation pattern which would occur for an omni-directional base station antenna and free space propagation. Note of course that the actual cellular footprint is determined by the contour in which a given transmitter serves the mobiles successfully. When using hexagons to model coverage areas, base station transmitters are depicted as either being in the center of the cell, center-excited cells or on three of the six cell vertices, edge-excited cells. Normally, omnidirectional antennas are used in center-excited cells, and sectored directional antennas are used in corner-excited cells. Practical considerations usually do not allow base stations to be placed exactly as they appear in the hexagonal layout. Most system designs permit a base station to be positioned up to one-fourth the cell radius away from the ideal location.

Figure 5.1: Illustration of the cellular frequency reuse concept. Cells with the same letter use the same set of frequencies. A cell cluster is outlined in bold and replicated over the coverage area. In this example, the cluster size N is equal to seven, and the frequency reuse factor is 1/7. Analytical Model Consider a cellular system with the followings.  S duplex channels available for use Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

48

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

 Each cell is allocated a group of k channels (k < S) and  S channels are divided among N cells (into unique and disjoint channel groups, which each have the same number of channels). The total number of available radio channels can be expressed as S = k.N

(5.1)

The N cells which collectively use the complete set of available frequencies is called a cluster. If a cluster is replicated M times within the system, the total number of duplex channels, C can be used as a measure of capacity and is given by C = M.k.N = M.S

(5.2)

Note from equation (5.2) that the capacity of a cellular system is directly proportional to the number of times a cluster is replicated in a fixed service area. The factor N is called the cluster size and is typically equal to 4, 7, or 12. The value of N plays an important role in system design and performance. If the cluster size N is reduced while the cell size is kept constant, more clusters are required to cover a given area. And hence, more capacity (a larger value of C) is achieved. In addition, the value for N is a function of how much interference a mobile or base station can tolerate while maintaining a sufficient quality of communications. Note that a large cluster size indicates that the ratio between the cell radius (R) and the distance between co-channel cells (D) is large. Conversely, a small cluster size indicates that co-channel cells are located much closer together. From a design viewpoint, the smallest possible value of N is desirable in order to maximize capacity over a given coverage area, i.e. to maximize C. The frequency reuse factor of a cellular system is given by 1/N since each cell within a cluster is only assigned 1/N of the total available channels in the system. To design network without gaps between adjacent cells, the geometry of hexagons is such that the number of cells per cluster N can only have values which satisfy equation (5.3). N=i2+ij+j2

(5.3)

where i and j are non-negative integers, and i  j are called shift parameters. Note that in order to find the nearest co-channel neighbors of a particular cell, the followings steps should consider. Step 1 move i cells along any chain of hexagons, then Step 2 turn 60 degrees counter-clockwise, and finally Step 3 move j cells; the jth cell is the co-channel cell. or Step 1 move j cells along any chain of hexagons, then Step 2 turn 60 degrees clockwise, and finally Step 3 move i cells; the ith cell is the co-channel cell. This is illustrated in figure 5.2 for i = 3 and j = 2, hence N = 19; an example illustration.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

49

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 5.2: Method of locating co-channel cells in a cellular system (for i = 3 and j = 2, hence N= 19). 5.2 CHANNEL ASSIGNMENT STRATEGIES A variety of channel assignment strategies have been developed to achieve the objectives: increasing capacity and minimizing interference for efficient utilization of the radio spectrum. Channel assignment strategies can be classified as either fixed or dynamic. Channel assignment strategy impacts the performance of the system, particularly during handing off a mobile phone from one cell to another. In a fixed channel assignment strategy,  Each cell is allocated a predetermined set of voice channels.  Any call attempt within the cell can only be served by the unused channels in that particular cell.  If all the channels in that cell are occupied, the call is blocked and the subscriber does not receive service. Note that several variations of the fixed assignment strategy exist such as borrowing strategy where a cell is allowed to borrow channels from a neighboring cell if all of its own channels are already occupied. The mobile switching center (MSC) supervises such borrowing procedures and ensures that the borrowing of a channel does not disrupt or interfere with any of the calls in progress in the donor cell. In a dynamic channel assignment strategy,  Voice channels are not allocated to different cells permanently. Instead, each time a call request is made, the serving base station requests a channel from the MSC.  The switch then allocates a channel to the requested cell following an algorithm that takes into account the likelihood of future blocking within the cell, the frequency of use of the candidate channel, the reuse distance of the channel, and other cost functions.  Accordingly, the MSC only allocates a given frequency if that frequency is not presently in use in the cell or any other cell which falls within the minimum restricted distance of frequency reuse to avoid co-channel interference. Dynamic channel assignment reduces the likelihood of blocking which increases the trunking capacity of the system since all the available channels in a market are accessible to all of the cells. However, dynamic channel assignment strategies require the MSC to collect real-time data on channel occupancy, traffic distribution, and radio signal strength indications (RSSI) of all channels on a continuous basis. This increases the storage and computational load on the system. The following table shows the assignment of traffic channels for the first half (Block A) of the 666 channel AMPS system (where 42 channels are reserved as control channels) for the frequency reuse factor K = 7 and 3 sectors per cell. It can be easily seen that in any group the minimum frequency separation between channels is 21.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

50

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The following figure shows a channel assignment corresponding to the table, when each 2π/3 sector antenna is located in the corner of hexagonal cell. 51

5.3 HANDOFF STRATEGIES When a mobile moves into a different cell while a conversation is in progress, the MSC automatically transfers the call to a new channel belonging to the new base station. This operation is called handoff that not only involves identifying a new base station, but also requires that the voice and control signals be allocated to channels associated with the new base station. Handoffs must be performed successfully and as infrequently as possible, and be imperceptible to the users. In order to meet these requirements, system designers must specify an optimum signal level at which to initiate a handoff. Once a particular signal level is specified as the minimum usable signal for acceptable voice quality at the base station receiver (normally taken as between -90 dBm and -100 dBm), a slightly stronger signal level is used as a threshold at which a handoff is made. This margin is given by   Pr handoff  Pr mini mum usable

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

and cannot be too large or too small. If  is too large, unnecessary handoffs which burden the MSC may occur, and if  is too small, there may be insufficient time to complete a handoff before a call is lost due to weak signal conditions. Figure 5.3 illustrates a handoff situation. Figure 5.3(a) demonstrates the case where a handoff is not made, and the signal drops below the minimum acceptable level to keep the channel active. This dropped call event can happen when there is an excessive delay by the MSC in assigning a handoff, or when the threshold  is set too small for the handoff time in the system.

52

Figure 5.3: Illustration of a handoff scenario at cell boundary. In deciding when to handoff, it is important to ensure that the drop in the measured signal level is not due to momentary fading, and that the mobile is actually moving away from the serving base station. In order to ensure this, the base station monitors the signal level for a certain period of time before a handoff is initiated. This running average measurement of signal strength should be optimized so that unnecessary handoffs are avoided, while ensuring that necessary handoffs are completed before a call is terminated due to poor signal level. The length of time needed to decide if a handoff is necessary depends on the speed at which the vehicle is moving. If the slope of the short-term average received signal level in a given time interval is steep, the handoff should be made quickly. Information about the vehicle speed which can be useful in handoff decisions can also be computed from the statistics of the received short-term fading signal at the base station. The time over which a call may be maintained within a cell, without handoff, is called the dwell time. The dwell time of a particular user is governed by a number of factors, which include propagation, interference, distance between the subscriber and the base station, and other time varying effects. Note that even when a mobile user is stationary, ambient motion in the vicinity of the base station and the mobile can produce fading, thus even a stationary subscriber may have a random and finite dwell time. It is apparent that the statistics of dwell time are important in the practical design of handoff algorithms. In first generation analog cellular systems, signal strength measurements are made by the base stations and supervised by the MSC. Each base station constantly monitors the signal strengths of all of its reverse voice channels to determine the relative location of each mobile user with respect to the base station tower. In addition to measuring the RSSI of calls in progress within the cell, a spare receiver in each base station, called the locator receiver, is used to determine signal strengths of mobile users which are in neighboring cells. The locator receiver is controlled by the MSC and is used to monitor the signal strength of users in neighboring cells which appear to be in need of handoff and reports all RSSI values to the MSC. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Based on the locator receiver signal strength information from each base station, the MSC decides if a handoff is necessary or not. In second generation systems that use digital TDMA technology, handoff decisions are mobile assisted. In mobile assisted handoff (MAHO), every mobile station measures the received power from surrounding base stations and continually reports the results of these measurements to the serving base station. A handoff is initiated when the power received from the base station of a neighboring cell begins to exceed the power received from the current base station by a certain level or for a certain period of time. The MAHO method enables the call to be handed over between base stations at a much faster rate than in first generation analog systems since the handoff measurements are made by each mobile, and the MSC no longer constantly monitors signal strengths. Different systems have different policies and methods for managing handoff requests. However, from the user’s point of view, having a call abruptly terminated while in the middle of a conversation is more annoying than being blocked occasionally on a new call attempt. To improve the quality of service as perceived by the users, various methods have been devised to prioritize handoff requests over call initiation requests when allocating voice channels. Such methods are as guard channel concept: a fraction of the total available channels in a cell is reserved exclusively for handoff requests from ongoing calls which may be handed off into the cell; queuing of handoff requests: decrease the probability of forced termination of a call due to lack of available channels. Practical Handoff Considerations 



In practical cellular systems, several problems arise when attempting to design for a wide range of mobile velocities. High speed vehicles pass through the coverage region of a cell within a matter of seconds, whereas pedestrian users may never need a handoff during a call. Several schemes have been devised to handle the simultaneous traffic of high speed and low speed users while minimizing the handoff intervention from the MSC. Another practical limitation is the ability to obtain new cell sites. Although the cellular concept clearly provides additional capacity through the addition of cell sites, in practice it is difficult for cellular service providers to obtain new physical cell site locations in urban areas. Zoning laws, ordinances, and other nontechnical barriers often make it more attractive for a cellular provider to install additional channels and base stations at the same physical location of an existing cell, rather than find new site locations. By using different antenna heights (often on the same building or tower) and different power levels, it is possible to provide "large" and "small" cells which are co-located at a single location. This technique is called the umbrella cell approach and is used to provide large area coverage to high speed users while providing small area coverage to users traveling at low speeds. Figure 5.4 illustrates an umbrella cell which is co-located with some smaller microcells. If a high speed user in the large umbrella cell is approaching the base station, and its velocity is rapidly decreasing, the base station may decide to hand the user into the co-located microcell, without MSC intervention.

Figure 5.4: The umbrella cell approach.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

53

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



Another practical handoff problem in microcell systems is known as cell dragging. Cell dragging results from pedestrian users that provide a very strong signal to the base station. Such a situation occurs in an urban environment when there is a line-of-sight (LOS) radio path between the subscriber and the base station. As the user travels away from the base station at a very slow speed, the average signal strength does not decay rapidly. Even when the user has traveled well beyond the designed range of the cell, the received signal at the base station may be above the handoff threshold, thus a handoff may not be made. This creates a potential interference and traffic management problem, since the user has mean while traveled deep within a neighboring cell. To solve the cell dragging problem, handoff thresholds and radio coverage parameters must be adjusted carefully.

In first generation analog cellular systems, the typical time to make a handoff, once the signal level is deemed to be below the handoff threshold, is about 10 seconds. This requires that the value for  be on the order of 6 dB to 12 dB. In new digital cellular systems such as GSM, the mobile assists with the handoff procedure by determining the best handoff candidates and the handoff once the decision is made, typically requires only 1 or 2 seconds. Consequently,  is usually between 0 dB and 6 dB in modern cellular systems. The IS-95 code division multiple access (CDMA) spread spectrum cellular system provides a unique handoff capability that cannot be provided with other wireless systems. Unlike channelized wireless systems that assign different radio channels during a handoff (called a hard handoff), spread spectrum mobiles share the same channel in every cell. Thus, the term handoff does not mean a physical change in the assigned channel, rather is that a different base station handles the radio communication task. By simultaneously evaluating the received signals from a single subscriber at several neighboring base stations, the MSC may actually decide which version of the user's signal is best at any moment in time. This technique exploits macroscopic space diversity provided by the different physical locations of the base stations and allows the MSC to make a soft decision as to which version of the user's signal to pass along to the PSTN at any instance. The ability to select between the instantaneous received signals from a variety of base stations is called soft handoff. 5.4 INTERFERENCE IN CELLULAR SYSTEMS Interference is the major limiting factor in the performance of cellular radio systems. Sources of interference include another mobile in the same cell, a call in progress in a neighboring cell, other base stations operating in the same frequency band, or any non-cellular system which inadvertently leaks energy into the cellular frequency band.  Interference on voice channels causes cross talk, where the subscriber hears interference in the background due to an undesired transmission. On control channels, interference leads to missed and blocked calls due to errors in the digital signaling.  Interference is more severe in urban areas, due to the greater high frequency (HF) noise floor and the large number of base stations and mobiles.  Interference has been recognized as a major bottleneck in increasing capacity and is often responsible for dropped calls.  The two major types of system-generated cellular interference are co-channel interference and adjacent channel interference.  They are difficult to control in practice due to random propagation effects and more so is the interference due to out-of-band users, which arises without warning due to front end overload of subscriber equipment or intermittent inter-modulation products. In practice, the transmitters from competing cellular carriers are often a significant source of out-of-band interference, since competitors often locate their base stations in close proximity to one another in order to provide comparable coverage to customers.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

54

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Co-channel Interference Frequency reuse implies that in a given coverage area, there are several cells that use the same set of frequencies. These cells are called co-channel cells, and the interference between signals from these cells is called co-channel interference. Note that unlike thermal noise which can be overcome by increasing the signal-to-noise ratio (SNR), co-channel interference cannot be combated by simply increasing the carrier power of a transmitter. This is because an increase in carrier transmit power increases the interference to neighboring co-channel cells. To reduce co-channel interference, co-channel cells must be physically separated by a minimum distance to provide sufficient isolation due to propagation. Consider that the size of each cell is approximately the same, and all the base stations transmit the same power. The co-channel interference then becomes a function of the radius of the cell (R) and the distance between centers of the nearest co-channel cells (D) and can be reduced by increasing the ratio of D/R. The parameter Q called the co-channel reuse ratio is related to the cluster size N, for a hexagonal geometry by the following. Q 

D



(5.4)

3N

R

Note that a small value of Q provides larger capacity since the cluster size N is small, whereas a large value of Q improves the transmission quality due to a smaller level of co-channel interference. Hence, a trade-off must be made between these two objectives in actual cellular design. Table 5.1: Co-channel Reuse Ratio for Some Values of N. 55

Let i0 be the number of co-channel interfering cells. Then, the signal-to-interference ratio (S/I or SIR) for a mobile receiver, which monitors a forward channel, can be expressed as S I



S i0  Ii i 1

(5.5)

where S is the desired signal power from the desired base station and Ii is the interference power caused by the ith interfering co-channel cell base station. The average received power Pr at a distance d from the transmitting antenna is approximated by  d   Pr  P 0  d   0 

n

(5.6)

where P0 is the power received at a close-in reference point in the far field region of the antenna at a small distance d0 from the transmitting antenna, and n is the path loss exponent. Now consider the forward link where the desired signal is the serving base station, and where the interference is due to co-channel base stations. If Di is the distance of the ith interferer from the mobile, the received power at a given mobile due to the ith interfering cell will be proportional to (Di)-n. Hence, S/I for a mobile can be approximated as (using equation 5.5) S I



R

n

i0  i 1

(5.7)

D i   n

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Considering only the first layer of interfering cells, if all the interfering base stations are equidistant from the desired base station, and if this distance is equal to the distance D between cell centers, then equation (5.7) simplifies to S

D R n   

I

i0

3N

n

(5.8)

i0

Note that equation (5.8) is based on the hexagonal cell geometry where all the interfering cells are equidistant from the base station receiver, and hence provides an optimistic result in many cases. For some frequency reuse plans (e.g. N = 4), the closest interfering cells vary widely in their distances from the desired cell.

56

Figure 5.5: Illustration of the first tier of co-channel cells for a cluster size of N=7. When the mobile is at the cell boundary (point A), it experiences worst case co-channel interference on the forward channel. The marked distances between the mobile and different co-channel cells are based on approximations made for easy analysis. From figure 5.5, it can be seen for a 7-cell cluster, with the mobile unit is at the cell boundary, the mobile is a distance D — R from the two nearest co-channel interfering cells and approximately D+R, D, D, and D+R from the other interfering cells in the first tier [Lee86]. Using equation (2.9) and assuming n equals 4, the signal-to-interference ratio for the worst case can be closely approximated as S



I

Or

S I

R 2 D  R 



4

4

 2 D  R 

4

 2D

4

1 2 Q  1 

4

 2 Q  1 

4

 2Q

4

(5.9)

(5.10)

To design the cellular system for proper performance in the worst case, it would be necessary to increase N to the next largest size. However, this obviously entails a significant decrease in capacity, since for example 12-cell reuse offers a spectrum utilization of 1/12 within each cell, whereas 7-cell reuse offers a spectrum utilization of 1/7. In practice, a capacity reduction of 7/12 would not be tolerable to accommodate for the worst case situation which rarely occurs. Hence, co-channel interference determines link performance, which in turn dictates the frequency reuse plan and the overall capacity of cellular systems. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Adjacent Channel Interference Interference resulting from signals which are adjacent in frequency to the desired signal is called adjacent channel interference. Adjacent channel interference results from imperfect receiver filters which allow nearby frequencies to leak into the passband. The problem can be particularly serious if an adjacent channel user is transmitting in very close range to a subscriber's receiver, while the receiver attempts to receive a base station on the desired channel. This is referred to as the near-far effect, where a nearby transmitter (which may or may not be of the same type as that used by the cellular system) captures the receiver of the subscriber. Alternatively, the near-far effect occurs when a mobile close to a base station transmits on a channel close to one being used by a weak mobile. The base station may have difficulty in discriminating the desired mobile user from the bleedover caused by the close adjacent channel mobile.     

Adjacent channel interference can be minimized through careful filtering and channel assignments. Since each cell is given only a fraction of the available channels, a cell need not be assigned channels which are all adjacent in frequency. By keeping the frequency separation between each channel in a given cell as large as possible, the adjacent channel interference may be reduced considerably. Thus instead of assigning channels which form a contiguous band of frequencies within a particular cell, channels are allocated such that the frequency separation between channels in a given cell is maximized. By sequentially assigning successive channels in the frequency band to different cells, many channel allocation schemes are able to separate adjacent channels in a cell by as many as N channel bandwidths, where N is the cluster size. Some channel allocation schemes also prevent a secondary source of adjacent channel interference by avoiding the use of adjacent channels in neighboring cell sites.

Note however that if the frequency reuse factor is small, the separation between adjacent channels may not be sufficient to keep the adjacent channel interference level within tolerable limits. For example, if a mobile is 20 times as close to the base station as another mobile and has energy spill out of its passband, the signal-to-interference ratio for the weak mobile (before receiver filtering) is approximately S

  20 

n

(5.11)

I

In practice, each base station receiver is preceded by a high Q cavity filter in order to reject adjacent channel interference. 5.5 TRUNKING AND GRADE OF SERVICE Cellular radio systems rely on trunking to accommodate a large number of users in a limited radio spectrum. The concept of trunking allows a large number of users to share the relatively small number of channels in a cell by providing access to each user, on demand, from a pool of available channels. In a trunked radio system, each user is allocated a channel on a per call basis, and upon termination of the call, the previously occupied channel is immediately returned to the pool of available channels. There is a trade-off between the number of available telephone circuits and the likelihood of a particular user finding that no circuits are available during the peak calling time. In a trunked mobile radio system, when a particular user requests service and all of the radio channels are already in use, the user is blocked, or denied access to the system. In some systems, a queue may be used to hold the requesting users until a channel becomes available. Note that one Erlang represents the amount of traffic intensity carried by a channel that is completely occupied (i.e. 1 call-hour per hour or 1 call-minute per minute). For example, a radio channel that is occupied for thirty minutes during an hour carries 0.5 Erlangs of traffic. The grade of service (GOS) is a measure of the ability of a user to access a trunked system during the busiest hour. The busy hour is based upon customer demand at the busiest hour during a week, month, or year. The busy hours for cellular radio systems typically occur during rush hours, between 4 p.m. and 6 p.m. on a Thursday or Friday evening. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

57

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The grade of service is a benchmark used to define the desired performance of a particular trunked system by specifying a desired likelihood of a user obtaining channel access given a specific number of channels available in the system. GOS is typically given as the likelihood that a call is blocked, or the likelihood of a call experiencing a delay greater than a certain queuing time. A number of definitions listed in Table 5.3 are used in trunking theory to make capacity estimates in trunked systems. Table 5.3: Definitions of common terms used in trunking theory.

The traffic intensity offered by each user is equal to the call request rate multiplied by the holding time. That is, each user generates a traffic intensity of Au Erlangs given by Au   H

(5.12)

where H is the average duration of a call, and  is the average number of call requests per unit time. For a system containing U users and an unspecified number of channels, the total offered traffic intensity A is given as A = UAu

(5.13)

Furthermore, in a C channel trunked system, if the traffic is equally distributed among the channels, then the traffic intensity per channel Ac is given as Ac 

UA u

(5.14)

C

Note that the offered traffic is not necessarily the traffic which is carried by the trunked system, only that which is offered to the trunked system. When the offered traffic exceeds the maximum capacity of the system, the carried traffic becomes limited due to the limited capacity (i.e. limited number of channels). The maximum possible carried traffic is the total number of channels, C, in Erlangs. The AMPS cellular system is designed for a GOS of 2% blocking. This implies that the channel allocations for cell sites are designed so that 2 out of 100 calls will be blocked due to channel occupancy during the busiest hour. There are two types of trunked systems which are commonly used. The first type offers no queuing for call requests. That is, for every user who requests service, it is assumed there is no setup time and the user is given immediate access to a channel if one is available. If no channels are available, the requesting user is blocked without access and is free to try again later. This type of trunking is called blocked calls cleared and assumes that calls arrive as determined by a Poisson distribution. Furthermore, it is assumed that there are an infinite number of users as well as the following:  there are memoryless arrivals of requests, implying that all users, including blocked users, may request a channel at any time Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

58

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

 

the probability of a user occupying a channel is exponentially distributed, so that longer calls are less likely to occur as described by an exponential distribution, and there are a finite number of channels available in the trunking pool. This is known as an M/M/m queue, and leads to the derivation of the Erlang B formula (also known as the blocked calls cleared formula).

The Erlang B formula determines the probability that a call is blocked and is a measure of the GOS for a trunked system which provides no queuing for blocked calls. The Erlang B formula is given by A

C

C!

Pr( call blocking ) 

C  k 0

A

k

 GO S

(5.15)

k!

where C is the number of trunked channels offered by a trunked radio system and A is the total offered traffic. Note that the Erlang B formula provides a conservative estimate of the GOS, as the finite user results always predict a smaller likelihood of blocking. The second kind of trunked system is one in which a queue is provided to hold calls which are blocked. If a channel is not available immediately, the call request may be delayed until a channel becomes available. This type of trunking is called Blocked Calls Delayed, and its measure of GOS is defined as the probability that a call is blocked after waiting a specific length of time in the queue.To find the GOS, it is first necessary to find the likelihood that a call is initially denied access to the system. The likelihood of a call not having immediate access to a channel is determined by the Erlang C formula is given by A

Pr( delay  0 )  A

C

C

A    C!  1   C  

C 1  k 0

A

k

(5.16)

k!

If no channels are immediately available the call is delayed, and the probability that the delayed call is forced to wait more than t seconds is given by the probability that a call is delayed, multiplied by the conditional probability that the delay is greater than t seconds. The GOS of a trunked system where blocked calls are delayed is hence given by Pr( delay  t )  Pr delay  0  Pr delay  t delay  0 

(5.17)

t    Pr delay  0  exp   C  A   H  

The average delay D for all calls in a queued system is given by D  Pr delay  0 

Note here that

H C A

H C A

(5.18)

is the average delay for those calls which are queued.

The Erlang B and Erlang C formulas are plotted in graphical form in figure 5.6 and figure 5.7. These graphs are useful for determining GOS in rapid fashion. To use figure 5.6 and figure 5.7, locate the number of channels on the top portion of the graph. Locate the traffic intensity of the system on the bottom portion of the graph. The blocking probability Pr (blocking) is shown on the abscissa of figure 5.6, and Pr (delay >0) is shown on the abscissa of figure 5.7. With two of the parameters specified it is easy to find the third parameter. Trunking efficiency is a measure of the number of users which can be offered a particular GOS with a particular configuration of fixed channels. The way in which channels are grouped can substantially alter the number of users handled by a trunked system. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

59

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

For example, from table 5.4, 10 trunked channels at a GOS of 0.01 can support 4.46 Erlangs of traffic, whereas 2 groups of 5 trunked channels can support 2 x 1.36 Erlangs, or 2.72 Erlangs of traffic. Clearly, 10 channels trunked together support 60% more traffic at a specific GOS than do two 5 channel trunks. It should be clear that the allocation of channels in a trunked radio system has a major impact on overall system capacity. Table 5.4: Capacity of an Erlang B System.

60

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

61

Figure 5.6: The Erlang B chart showing the probability of blocking as functions of the number of channels and traffic intensity in Erlangs.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

62

Figure 5.7: The Erlang C chart showing the probability ofa call being delayed as a function of the number of channels and traffic intensity in Erlangs. 5.6 IMPROVING CAPACITY IN CELLULAR SYSTEMS Techniques such as cell splitting, sectoring, and coverage zone approaches are used in practice to expand the capacity of cellular systems. Cell splitting allows an orderly growth of the cellular system. Sectoring uses directional antennas to further control the interference and frequency reuse of channels. The zone microcell concept distributes the coverage of a cell and extends the cell boundary to hard-to-reach places. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

While cell splitting increases the number of base stations in order to increase capacity, sectoring and zone microcells rely on base station antenna placements to improve capacity by reducing co-channel interference. Cell splitting and zone microcell techniques do not suffer the trunking inefficiencies experienced by sectored cells, and enable the base station to oversee all handoff chores related to the microcells and thus reducing the computational load at the MSC. Cell Splitting Cell splitting is the process of subdividing a congested cell into smaller cells, each with its own base station and a corresponding reduction in antenna height and transmitter power. Cell splitting increases the capacity of a cellular system since it increases the number of times that channels are reused. By defining new cells which have a smaller radius than the original cells and by installing these smaller cells between the existing cells, capacity increases due to the additional number of channels per unit area. An example of cell splitting is shown in figure 5.8. In figure 5.8, the base stations are placed at corners of the cells, and the area served by base station A is assumed to be saturated with traffic (i.e., the blocking of base station A exceeds acceptable rates). New base stations are therefore needed in the region to increase the number of channels in the area and to reduce the area served by the single base station. Note in the figure that the original base station A has been surrounded by six new microcell base stations. In the example shown in figure 5.8, the smaller cells were added in such a way as to preserve the frequency reuse plan of the system. For example, the new cell base station labeled G was placed half way between two larger stations utilizing the same channel set G. This is also the case for the other new cells in the figure. As can be seen from figure 5.8, cell splitting merely scales the geometry of the cluster. In this case, the radius of each new microcell is half that of the original cell. For the new cells to be smaller in size, the transmit power of these cells must be reduced. The transmit power of the new cells with radius half that of the original cells can be found by examining the received power at the new and old cell boundaries and setting them equal to each other. This is necessary to ensure that the frequency reuse plan for the new microcells behaves exactly as for the original cells. For figure 5.8

Figure 5.8: Illustration of cell splitting. P r at old cell boundary

Pr at new cell boundary

  P t1 R  n

  Pt 2 ( R / 2 )  n

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

(5.18) and (5.19),

edition 02, version 02 2013

63

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

where Pt1 and Pt2 are the transmit powers of the larger and smaller cell base stations, respectively, and n is the path loss exponent. If we take n = 4 and set the received powers equal to each other, then Pt 2 

P t1

(5.20).

16

In other words, the transmit power must be reduced by 12 dB (-4*10log10 (1/2) dB) in order to fill in the original coverage area with microcells, while maintaining the S/I requirement. In practice, not all cells are split at the same time. It is often difficult for service providers to find real estate that is perfectly situated for cell splitting. Therefore, different cell sizes will exist simultaneously. In such situations, special care needs to be taken to keep the distance between co-channel cells at the required minimum, and hence channel assignments become more complicated. When there are two cell sizes in the same region as shown in figure 5.8, one cannot simply use the original transmit power for all new cells or the new transmit power for all the original cells. If the larger transmit power is used for all cells, some channels used by the smaller cells would not be sufficiently separated from co-channel cells. On the other hand, if the smaller transmit power is used for all the cells, there would be parts of the larger cells left unserved. For this reason, channels in the old cell must be broken down into two channel groups, one that corresponds to the smaller cell reuse requirements and the other that corresponds to the larger cell reuse requirements. The larger cell is usually dedicated to high speed traffic so that handoffs occur less frequently. At the beginning of the cell splitting process, there will be fewer channels in the small power groups. However, as demand grows, more channels will be required, and thus the smaller groups will require more channels. This splitting process continues until all the channels in an area are used in the lower power group, at which point cell splitting is complete within the region, and the entire system is rescaled to have a smaller radius per cell. Antenna down tilting (oriented), which deliberately focuses radiated energy from the base station towards the ground rather than towards the horizon (sphere), is often used to limit the radio coverage of newly formed microcells. Sectoring Cell splitting achieves capacity improvement by essentially rescaling the system. By decreasing the cell radius R and keeping the co-channel reuse ratio D/R unchanged, cell splitting increases the number of' channels per unit area. However, another way to increase capacity is to keep the cell radius unchanged and seek methods to decrease the D/R ratio. In this approach, capacity improvement is achieved by reducing the number of cells in a cluster and thus increasing the frequency reuse. However, in order to do this, it is necessary to reduce the relative interference without decreasing the transmit power. The co-channel interference in a cellular system may be decreased by replacing a single omnidirectional antenna at the base station by several directional antennas, each radiating within a specified sector. By using directional antennas, a given cell will receive interference and transmit with only a fraction of the available co-channel cells. The technique for decreasing co-channel interference and thus increasing system capacity by using directional antennas is called sectoring. The factor by which the co-channel interference is reduced depends on the amount of sectoring used. A cell is normally partitioned into three 1200 sectors or six 600 sectors as shown in figure 5.9 (a) and (b). When sectoring is employed, the channels used in a particular cell are broken down into sectored groups and are used only within a particular sector, as illustrated in figure 5.10 (a) and (b). Assuming 7cell reuse, for the case of 120° sectors, the number of interferers in the first tier is reduced from 6 to 2. This is because only 2 of the 6 co-channel cells receive interference with a particular sectored channel group. Referring to figure 5.10, consider the interference experienced by a mobile located in the right-most sector in the center cell labeled “5”. There are 3 co-channel cell sectors labeled "5" to the right of the center cell, and 3 to the left of the center cell. Out of these 6 co-channel cells, only 2 cells have sectors with antenna patterns which radiate into the center cell, and hence a mobile in the center cell will experience interference on the forward link from only these two sectors. The resulting S/I for this case can be found to be 24.2 dB, which is a significant improvement over the omni-directional case in, where the worst case S/I was 17 dB. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

64

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

In practical systems, further Improvement in S/I is achieved by downtilting the sector antennas such that the radiation pattern in the vertical (elevation) plane has a notch at the nearest co-channel cell distance. Thus, sectoring reduces interference, which amounts to an increase in capacity by a factor of 12/7 or 1.714. In practice, the reduction in interference offered by sectoring enable planners to reduce the cluster size N and provides an additional degree of freedom in assigning channels. The penalty for improved S/I and the resulting capacity improvement is an increased number of antennas at each base station, and a decrease in trunking efficiency due to channel sectoring at the base station. Since sectoring reduces the coverage area of a particular group of channels, the number of handoffs increases, as well. Fortunately, many modern base stations support sectorization and allow mobiles to be handed off from sector to sector within the same cell without intervention from the MSC, so the handoff problem is often not a major concern. It is the loss of traffic due to decreased trunking efficiency that causes some operators to shy away from the sectoring approach, particularly in dense urban areas where the directional antenna patterns are somewhat ineffective in controlling radio propagation. Because sectoring uses more than one antenna per base station, the available channels in the cell must be subdivided and dedicated to a specific antenna. This breaks up the available trunked channel pool into several smaller pools, and decreases trunking efficiency.

65

Figure 5.9: Sectorization of cells: (a) 1200 sectoring (b) 600 sectoring.

Figure 5.10: Illustration of how 1200 sectoring reduces interference from co-channel cells. Out of the 6 cochannel cells in the first tier, only 2 of them interfere with the center cell. If omni-directional antennas were used at each base station, all 6 co-channel cells would interfere with the center cell. Microcell Zone Concept

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The increased number of handoffs required when sectoring is employed results in an increased load on the switching and control link elements of the mobile system. A solution to this problem was presented by Lee. This proposal is based on a microcell concept for 7 cell reuse, as illustrated in figure 5.11. In this scheme, each of the three (or possibly more) zone sites (represented as Tx/Rx in figure 5.11) is connected to a single base station and share the same radio equipment. The zones are connected by coaxial cable, fiber optic cable, or microwave link to the base station. Multiple zones and a single base station make up a cell. As a mobile travels within the cell, it is served by the zone with the strongest signal. This approach is superior to sectoring since antennas are placed at the outer edges of the cell, and any base station channel may be assigned to any zone by the base station. As a mobile travels from one zone to another within the cell, it retains the same channel. Thus, unlike in sectoring, a handoff is not required at the MSC when the mobile travels between zones within the cell. The base station simply switches the channel to a different zone site. In this way, a given channel is active only in the particular zone in which the mobile is traveling, and hence the base station radiation is localized and interference is reduced. The channels are distributed in time and space by all three zones and are also reused in co-channel cells in the normal fashion. This technique is particularly useful along highways or along urban traffic corridors. The advantage of the zone cell technique is that while the cell maintains a particular coverage radius, the co-channel interference in the cellular system is reduced since a large central base station is replaced by several lower powered transmitters (zone transmitters) on the edges of the cell. Decreased cochannel interference improves the signal quality and also leads to an increase in capacity, without the degradation in trunking efficiency caused by sectoring. Zone cell architectures are being adopted in many cellular and personal communication systems. 66

Figure 5.11: The microcell concept. 5.7 SUMMARY In this chapter, the fundamental concepts of handoff, frequency reuse, trunking efficiency, and frequency planning have been presented. Handoffs are required to pass mobile traffic from cell to cell, and there are various ways handoffs are implemented. The capacity of a cellular system is a function of many variables. The S/I limits the frequency reuse factor of a system, which limits the number of channels within the coverage area. The trunking efficiency limits the number of users that can access a trunked radio system. Trunking is affected by the number of available channels and how they are partitioned in a trunked cellular Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

system. Trunking efficiency is quantified by the GOS. Finally, cell splitting, sectoring, and the zone microcell technique are all shown to improve capacity by increasing S/I in some fashion. All most all the information can be found in T. S. Rappaport, 2002, Wireless communications: Principles and Practice, Pearson Education LTD.

5.8 AN EXAMPLE CELLULAR SYSTEM DESIGN Analytical Derivation of Frequency Reuse Factor   

The frequency reuse is one of the most important design challenges in cellular networks. The base stations (BSs) of the cellular network should be located as near to each other as possible to maximize the number of simultaneous users, but should not be so close that the co-channel interference results in unacceptable speech quality. The reuse distance, i.e. distance between co-channel cells D is an important parameter that determines the reuse pattern within the cellular network.

Specifically, we assume that      

all co-channel base stations have same transmitter power, the path loss is proportional to d-4, noise level is negligible when compared to interference level there are 6 equidistant interferers at the channel, and the interference from interferers, that are further away, can be neglected, cell radius is R, and the distance between co-channel cells is D.

We can approximately analyze the co-channel interference on basis of the following figure

Hence, the signal to co-channel interference ratio is given as C

C

=

I

6

∑I i i =1

=

R 6D

4 4

so that Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

67

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

D  R4 6

C I

Note that the power of the transmitter does not occur in the preceding equation since “As long as the transmitted power at every co-channel cell site is the same, any transmitter power can be used without changing the signal to co-channel interference ratio”. Note that the application of the previous equation for the AMPS system, where acceptable speech quality can be obtained with a C/I ratio of 18 dB with 30 kHz channel bandwidth. And, this requirement can be converted into a relation between the cell radius R and the minimum distance between co-channel cells D using the previous formula and can easily calculate that the C/I ratio of 18 dB corresponds to requirement D > 4.4R. The ratio of co-channel cell site distance to the cell radius R is given by D

3K

=

R

For K=7 and i=2 and j=1, the desired carrier-to-interference ratio at the cell site is given by C C C  6  IT 6I ∑I i i 1



1

R

6 D

-n

R

1D  R    6 R 

68

≥ 18 dB

-n

n

≥ 18 dB

Considering the worst case8 of equality, we can write D R

D R

1

1 = (6 × 10

1 .8

)n

= 5 . 4 for n=4, the worst case scenario (40 dB/decade)

path loss means value of radio propagation exponent n=4). D  R , so that

D

D

 R D

 4 .4

in the normal case.

R

We can then express K as,

8

Note the situation that the interfering mobile units are on the nearest fringe of the neighboring six co-channel zones transmitting simultaneously can be regarded as the worst case.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

1   K   6 C  I T  3  

1 n

  1 

2

Note that “for low value of CIR, K is reduced. Hence, larger number of channels per cell and a large traffic capacity”. For n=4, and CIR=18 dB, K  10 Neglecting 1, for a higher CIR and n = 4, K 

2 C    3  I T 

Hence if CIR increases that results in decrease in co-channel interference and so results higher value of K. If Nf is the number of channels per cell, then N

f



Ba KB c Ba

 Bc

2 C    3  I T 

69

For CIR of 18 dB, Nf = 64. Here, Ba is the total available bandwidth in transmit and receive directions (12.5 MHz + 12.5 MHz in AMPS) and Bc is channel bandwidth equals to 30 kHz in AMPS. Analytical Derivation of Capacity of Cellular Systems Typically the area of a cell is split into 4 new cells, which means that the radius of each new cell is one half of the radius of the original cell. Assume 40 dB/decade pathloss so that the power can be reduced by 12 dB. Thus, after n splitting, the new traffic capacity within the original area will be Tn  4 T0 n

and the power of the transmitters will be

Pn  dB   P0  12 n

Two relatively often used efficiency measures are the channel efficiency, which is defined as the maximum number of channels that can be provided within a given bandwidth, and spectral efficiency, which is defined as the maximum number of calls that can be served within a given area. Note that in cellular networks, spectral efficiency, not the channel efficiency, is the important parameter that we want to maximize. The spectral efficiency of cellular FDMA and TDMA systems can be expressed in terms of number of channels per cell as follows.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

( N

f



B tot

)

Bc Kc B tot Bc



2C    3  I  m in

where Btot is the total bandwidth of the system Bc is the (equivalent) bandwidth of a single channel (C/I)min is the minimum acceptable signal-to-interference ration K is the reuse factor Hence is shown that we can increase the spectral efficiency by either decreasing the channel bandwidth Bc or decreasing the C/I requirement. In FDMA systems, Bc the bandwidth of a single channel, and in TDMA systems Bc is the equivalent bandwidth of a single channel. For example for 200 kHz TDMA channel with 8 traffic channels in GSM system, the equivalent bandwidth is 200/8 = 25 kHz. Thus, when we take the analog AMPS system with 30 kHz channels and C/I requirement of 18 dB as a reference (capacity = 1), we can calculate the following relative capacities with respect to AMPS.

Hence, the American and Japanese TDMA systems outperform GSM in this comparison because of their lower-rate speech codec, which permits the use of narrower bandwidth in transmission. However, in analog cellular systems, the channel bandwidth Bc and the C/I requirement are interrelated. By approximately doubling the transmission bandwidth in frequency modulation, we can increase the output SNR at the receiver by 6 dB. Correspondingly, if we reduce the channel bandwidth to one half, we should increase the C/I requirement by 6 dB (by a factor of 4) for the same speech quality. For example, to maintain the same speech quality, when Bc is reduced from 30 to 15 kHz, the C/I requirement has to be increased from 18 to 24 dB. Note that it has been found out that roughly 60 channels per cell can be provided for an allocated bandwidth of 12.5 MHz in analog cellular systems, regardless of the bandwidth of individual channels, which can be 30, 15 or 5 kHz. We can calculate the bit transfer capacity: number of bits transmitted within one second in bandwidth of one Hertz in a single cell of the network. For GSM system with speech codec rate of 13 kbit/s, equivalent bandwidth of 200/8 kHz and cell reuse factor of 4, we get the result

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

70

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

c 

Rt Bc K

 13 kbps 

1



200

1 4

8 kHz  0 . 13 bps / Hz / cell

For North American digital TDMA system with speech codec rate of 8 kbit/s equivalent bandwidth of 30/3 kHz and cell reuse factor of 7, we get the result c 

Rt Bc K

 8 kbps 

1 30



1 7

3 kHz  0 . 11 bps / Hz / cell

For Japanese digital TDMA system with speech codec rate of 8 kbit/s, equivalent bandwidth of 25/3 kHz and cell reuse factor of 7, we get the result c 

Rt 71

Bc K  8 kbps 

1 25



1 7

3 kHz  0 . 14 bps / Hz / cell

If we assume that the speech quality of AMPS system corresponds to the speech codec rate of 10kbit/s, we can calculate the capacity for AMPS with bandwidth of 30 kHz and cell reuse factor of 7 as c 

Rt Bc K 1

 10 kbps 



30 kHz

1 7

 0 . 05 bps / Hz / cell

On the other hand, the designers of Qualcomm code-division multiple access (CDMA) system believe that they can approach the capacity c  1 bps / Hz / cell

However, this figure is based on utilizing discontinuous transmission with voice activity detection and antenna sectorization. In practice, due to difficulty of cellular planning in physical terrain and to irregular division of traffic, the spectrum efficiencies of cellular systems are only about half of the calculated ideal capacities. Traffic Channel Assignment Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

For an allocation of N frequency channels for a coverage area A, the capacity of the system can be assessed in terms of number of simultaneous calls, which can be calculated as c 

A  N    A BS  K 

where ABS is the area of a single base station. For hexagonal cells, this area can be calculated from the cell radius as A BS 

3

3R

2

2

 2 .6 R

2

For a C/I ratio of 18 dB a reuse factor K ≥ 7 are required. Now, for an allocation of N = 200 frequency channels for a coverage area A = 50 × 50km2 with the cell radius 5 km, the number of cells is N BS 

A

 38 . 5  38

A BS

and the number of simultaneous calls in the system is n  N BS  38

N K 200

 1085

72

7

Adjacent Channel Interference Adjacent channel interference occurs when signal energy spills over from one channel into another channel that is adjacent to it in the frequency spectrum. The most important adjacent channel interference is caused by the immediately preceding and following channels. In principle it is possible to control adjacent channel interference completely through filtering at the transmitter and the receiver. However, tight filtering makes mobile units more expensive and introduces ISI in the received signal. Assume a relatively simple receiving filter with 6 dB/oct low-pass equivalent attenuation characteristics, we can calculate for a 30 kHz channel the attenuation from the edge of the band at f1 = 15 kHz to frequency offset f2 as A  dB  

 f  log  2  log  2   f1  6

For f2 = 120 kHz and (f2/f1) = 8, i.e. 4 channels away, we get the attenuation of 18 dB. Thus, if we require C/I ratio of 18 dB as for cochannel interference, the three adjacent channels to any channel in use cannot be used for traffic. Adjacent channel interference can be reduced by maintaining maximum frequency separation between channels in any given cell. For maximum frequency separation between channels, the frequencies k, k+K, k+2K ... should be assigned to the kth cell, when the frequency reuse factor K is used. The following table shows the procedure of the frequency allocation used for the AMPS system for the frequency reuse factor K = 7.The letters A, B and C refer to 3 sectors of directional antennas that are used in each BS.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

   

Since the adjacent group numbers contain adjacent frequency channels, they should not be assigned to neighboring sectors in cells. It should be noted that for small reuse factors (e.g. K = 4), it is impossible (especially, without sectoring) to avoid using immediately adjacent frequency channels in some of the neighboring six cells. Already, for K = 7, it is very difficult without sectoring to find a channel assignment in which the use of immediately adjacent frequency channels could be avoided. However, for K = 9, we can find a channel assignment (see following Figure) in which immediately adjacent frequency channels are never used in neighboring cells.

73

An example of adjacent Channel frequency separation for two channels which are used in a 12-cell cluster is shown below. Here K=12, Ch. 1, 13, 25, 30 kHz is channel bandwidth center to center is 360 kHz Edge to edge is 330 kHz.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Frequency separation between desired and system-adjacent channel Note that all the preceding discussion has been describing fixed channel assignment, where each cell is permanently allocated a preselected set of frequency channels. Near-End-to-Far-End Ratio When the mobile units are moving within a cell, we often encounter a situation, where two MSs are transmitting simultaneously to the BS at different channels. And one of the MSs is much nearer to the BS than the other. For example, in a case the distances of MSs from the BS are 0.5 and 10 km and assuming 40 dB/dec pathloss, the ratio of received powers at the BS is 52 dB. This ratio of received powers due to different locations of two transmitters is called near-end-to-far-end ratio.

Assuming that the mobile transmits the same power, the signal received at the cell site is proportional to the geographical distance between the cell site and the MS. When the separation from two MS operating on adjacent channels is widely different, a situation can arise when the power received at the cell site from a nearby MS is far higher than that of from another MS farther away. These unbalanced received powers are due to different path losses and known as near-end (NE) to farend (FE) ratio interference (NE / FE) = (path loss due to path d1) / (path loss due to path d2), where nearby distance d1 is from interfering MS and farther distance d2 is from the desired MS. If n=4, then Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

74

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

 d    1  FE  d2 

4

NE

This large imbalance in the received levels makes the problem of adjacent channel interference even more severe, and as a consequence it also increases the required frequency separation between channels.  If we still require C/I ratio of 18 dB, the required attenuation of the lowpass-equivalent receiving filter for the lower-level channel is 52 + 18 = 70 dB.  Even if we assume a slightly better receiving filter with 12 dB/oct attenuation characteristics, the required frequency ratio f2/f1 is 58.  Even with more realistic 24 dB/oct attenuation characteristics, the required frequency ratio is still 8. This means that a separation of 29 or 4 channels is needed. For 30 kHz channels this means frequency separation of 870 or 120 kHz. An effective method for reducing effects of increased adjacent channel interference due to the high nearend-to-far-end ratio is the power control for MSs, where each MS transmits at the minimum power level that still ensures reliable communication.

75

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 06 GLOBAL SYSTEM FOR MOBILE COMMUNICATIONS 6.1 DIGITAL CELLULAR RADIO SYSTEMS The second generation cellular mobile systems are digital since in addition to signaling, speech signals also are encoded into digital form prior to transmission. The main service is the transmission of speech signals or mobile telephony. The purely digital nature of these systems enables a wide variety of services that were beyond the first generation of cellular mobile systems. In one system, users access the system by means of time division multiple access (TDMA), which means frequency channels are divided into time slots that are divided among users. The other system uses code division multiple access (CDMA), in which the users use all available spectrum simultaneously, but in such a way that each user is distinguished from all others by a unique code imprinted upon his transmitted signal. However, only the basic channel (bandwidth of 200 kHz in the GSM, 30 kHz in the IS-54, 25 kHz in the JDC, and 1250 kHz in the IS-95 system) is divided among users by using TDMA or CDMA. Multiples of these basic channel bandwidths are always divided among users by using frequency division. The currently operating second generation cellular mobile systems are as follows.  European Global System for Mobile communications (GSM)  North American digital cellular TDMA system defined in the U.S. standard IS-54  Japanese digital cellular TDMA system (JDC)  North American digital cellular CDMA system defined in the U.S. standard IS-95. A summary of technical characteristics of the existing and proposed digital cellular mobile systems is shown in the following table 7.1. 76

6.2 GSM OVERVIEW In 1982, the Conference of European Posts and Telegraphs (CEPT) formed a study group called the Groupe Spécial Mobile (GSM) to study and develop a pan-European public land mobile system. The proposed system had to meet certain criterion such as spectral efficiency, subjective speech quality, terminal and service cost, feasibility of handheld terminals, support for international roaming, support for new services, coexistence with existing systems, and ISDN compatibility. Commercial GSM service was started in mid-1991, and in 1992, European wide operation of GSM had already been rolled out. The first DCS1800 network became operational in 1993. Although standardized in Europe, GSM networks (including DCS1800 andPCS1900) are at present in operation around the world. With North America entering into the GSM world with a derivative of GSM called PCS1900, GSM systems now exist on every continent and the acronym GSM now aptly stands for Global System for Mobile communications.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 6.1: Technical characteristics of digital cellular mobile systems.

77

6.3 GSM SERVICES GSM is an integrated voice-data service in contrast to analog communications system where data service such as fax was defined as overlay services on top of the main (only single application) voice service. Services in GSM are divided into three categories such as Teleservices provide communication between two end user applications following certain standard protocol; Bearer services provide capabilities to transmit information among user-network-interfaces; and Supplementary services (which are not standalone services) supplement a bearer- or teleservice. Tables 6.2 and 6.3 show GSM phases 1 and phase 2 services. Table 6.2: GSM phase 1 services.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 6.3: GSM phase 2 additional services.

6.4 GSM ARCHITECTURE GSM was designed to be compatible with ISDN systems, and the services provided by GSM are a subset of the standard ISDN services (speech is the most basic). The functional architecture of a GSM system can be divided into the Mobile Station (MS), the Base Station (BS), and the Network Subsystem (NS). The MS is carried by the subscriber, the BS subsystem controls the radio link with the MS and the NS performs the switching of calls between the mobile and other fixed or mobile network users as well as mobility management. The MS and the BS subsystem communicate across the Um interface also known as radio link. The generic layout of GSM networks in terms of functional machines and interfaces is illustrated in figure 6.1.

(a) where

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

78

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

79

(b) Figure 6.1: (a) Generic layout of GSM network structure. (b) A different view of the reference architecture for GSM. The Mobile Station is the cellular phone carried by the subscriber. The Base Station Subsystem controls the radio link to the MS. The Switching Subsystem, the main part of which is the Mobile Switching Center, performs the switching of calls between a mobile user and other fixed or mobile network users, as well as mobility management. The Operation Subsystem oversees the proper operation and setup of the whole network. The Mobile Station (MS) represents the only piece of equipment that a user sees from the system. The MS includes the Terminal Equipment (TE), which carries out the functions specific to the service, and the Mobile Termination (MT), which carries out all functions related to the radio interface. If the interfaces of the TE and the MT are not compatible, the MS may also include a Terminal Adapter (TA). The subscriber identity module (SIM) is also considered a part of the MS since normally the MS cannot be operated without a SIM. The SIM is a smart card that contains all the subscriber-related information that is needed on the user's side of the radio interface. The insertion of SIM to a MS personalizes the MS so that actions of the MS can be associated with a specific subscriber (e.g. for charging). The SIM card may be protected against unauthorized use by a password or personal identity number. The mobile station is uniquely identified by its International Mobile Equipment Identity (IMEI). The SIM card contains the International Mobile Subscriber Identity (IMSI) used to identify the subscriber to the system. The IMEI and the IMSI are independent, thereby allowing users to utilize any GSM MS without being tied to a particular piece of hardware. The base station subsystem (BSS) contains the network infrastructure, which is specific to cellular radio aspects of GSM. The BSS includes the base transceiver station (BTS), which takes care of communication to the MSs through the radio interface, and the base station controller (BSC), which takes care of communication to the MSC through the A interface (an SS7 link). The BTS is responsible for providing an error-corrected data path over the radio interface. It is typical that several BTSs are located at the same site, sharing a common antenna tower and serving sectored cells.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The BSC is in charge of the radio interface management within a region, especially the allocation of radio channels and control of handovers. For each cell of the system a BTS is needed, but a BSC can control several (typically 20 to 30) BTSs. Within a switching subsystem (SSS), the MSC performs all the switching functions needed for the operation of the MSs within the group of cells that it services. The MSC acts like a normal switching node of the PSTN and additionally provides all the functionality needed to handle a mobile subscriber such as registration, location updating, handovers, and call routing to a roaming subscriber. The MSC takes also care of internetworking with PSTN and other networks (ISDN, circuit and packet switched data networks, etc.). An MSC controls several BSCs, and typically one MSC can serve a relatively large city of about 1 million inhabitants. Besides MSCs, the SSS includes several databases: HLR, VLR, AuC, and EIR. The home location register (HLR) contains all the data of each mobile subscriber registered in the GSM network of the associated MSC along with the current location of the subscriber. The permanent data in the HLR include IMSI, phone number, and permitted supplementary services. The temporary data in the HLR include the address of the current VLR which currently administrates the user, forwarding number in case of call forwarding and several transient parameters for authentication and encryption. The information about the current location of the subscriber (the address of the current VLR) allows the MSC to route the incoming calls to the MSC in command of the area, where the MS roams. There is logically one HLR per GSM network although it may be implemented as a distributed database. The authentication center (AuC) is defined as an independent module in SSS, but in practice, it is often implemented as a module within HLR. However, the logical separation of AuC and HLR is important since AuC is only used for security functions, especially the authentication of MSs. A SIM card issued in the area assigned to AuC contains the same authentication algorithm as the AuC does. The AuC issues an input and the corresponding output of the authentication algorithm to either HLR or VLR for authentication of subscribers. The visitor location register (VLR) is used for temporarily storing the information of mobile subscribers that are currently located in the service area of the corresponding MSC. Specifically, the VLR contains more accurate location of a subscriber than what is stored in the HLR of the subscriber and the temporary mobile subscriber identity (TMSI) which is used for limited periods of time to avoid the transmission of the IMSI over radio interface. In practical implementations the VLR is integrated as a part of each MSC so that the geographical area controlled by the MSC corresponds to that controlled by the VLR. The last database of SSS is the equipment identity register (EIR) that contains IMEI of all MSs that are known to be stolen or faulty or that have been used fraudulently. The IMEI identifies the manufacturer, the country of production, and the type approval of the MS. Figure 6.2 shows the interconnections in entities of GSM networks.

where

Figure 6.2: Interconnections in entities of GSM networks.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

80

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION 6.5 POSITIONING GSM TELEPHONE

Cellular positioning, phone tracking is the ability to locate every powered-on, authorized cell phone in a GSM/UMTS network. GSM implements the possibility of positioning a mobile device. In general case, this can be made with the IMSI number and the cell in which the wireless device is currently connected. It depends on several components such as the landscape, the signal strength, the utilization of the cell and the cell surface. There are several reasons for positioning such as given below. Location-Sensitive Billing – Communication providers can introduce tariffs depending on the position of cell phone. They can differ where the person is calling from, thus international calls can be charged higher than domestic calls (e.g. Roaming tariffs) Increased subscriber safety – An increasing number of emergency calls are made from mobile phones. In many cases the caller cannot exactly announce his position. In such case it is useful to determine his position automatically by a positioning system. Enhanced network Performance - At the microscopic level, accurately monitoring the movement of mobile telephones enables a cellular communications network to make better decisions on when to hand over from one cell to the next. Macroscopically, long-term monitoring of mobile telephone positions provides excellent input to the planning of the cellular network. However, on top of all, the most important application is increased subscriber safety, due to locating emergency calls. Types of positioning systems For the classification of a positioning system it is important where the positioning is made and where the information is used. There are two types of positioning systems such as self-positioning and remotepositioning. The classification is useful for analyzing an existent system. Self-positioning In a self-positioning, the receiver makes its signal measurement from geographically distributed transmitters and uses its results for determining the current position. A good example for a self-positioning system is a GPS receiver. It gets signal from several satellites and it compute the position. Remote-positioning In a remote-positioning system the receivers, located at different places, measure the signal, emitted by an object which has to be localized. The data is sent to a central server where the object’s position is estimated. The information can be used for different applications (e.g. CAD-Systems). Indirect-Positioning It is possible to transmit data from a self-positioning system to a central site or vice versa via data link. There a many methods to derive position of signal measurement, which can be applied to GSM. The most important parameters are propagation time, time difference of arrival (TDOA), angle of arrival (AOA) and carrier phase. The intersection of the measured values can be assumed as the current position of the mobile phone. A least squares approach of these values can be done for determining position with a minimal error. The more measurements are made, the more accurate the result. However, if too few values are available, the loci will intersect at more than one point, resulting in ambiguous position results. Figure 6.3 shows base functionality of a GSM positioning system that works with mentioned parameters propagation time, TDOA and AOA.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

81

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Figure 6.3: theoretical model of GSM positioning system with shown parameters; A, B, C represent base stations, X is the locus of the telephone. Positioning parameters Propagation time This involves measuring the time it takes for a signal to travel between a base station and a mobile telephone or vice versa. It can also be seen as the round- trip time of a signal: It is transmitted from a station to its destination and is echoed back. This time can differ from one location to the other. In case of movement of the caller, it is more difficult to accurately estimate his position, because of different frequencies of the signal. When an object is moving toward a base station, the emitted wavelength of the object’s signal is shorter than the current. When it moves away, the wavelength seems to be higher, caused by the Doppler Effect. Although there are mathematical models of describing this effect, but it would take too long to describe these models related to GSM positioning in here. Let’s take a look at the case figured in (Fig. 6.3 (a)). There are two different values measured. This leads to a circular view around the mobile phone. Each arrival of propagation time value can be seen around the phone because the angle of the incoming signal is not defined. Mathematically there two circles that intersect in two points. We see two positions that might be impossible. This case shows that positioning estimation with one or more propagation times is not efficient enough. Time difference of arrival (TDOA) A mobile phone can listen to several base stations and computes time difference between one pair of arrivals. For example, three values of three base stations are measured; two independent TDOA measurements can be made. Each TDOA measurement defines a hyperbolic locus, where the telephone must lie. The intersection of more values will define current position. (See Fig. 6.3 (b)) Angle of Arrival (AOA) The angle of arrival is the angle of incoming signal of the base station at angle a mobile phone or angle from telephone at base station. One measurement produces a straight line from base station to phone. This is no more meaningful. A second measurement will yield a second line, which can be intersected with the first one. The result is the telephone’s position. (See Fig. 6.3 (c)) Carrier phase Mobile phones can use the GSM carrier wave to compute their positions, but there are many problems using this parameter. The positioning receiver can measure the phase of the received signal, but it cannot count the cycles (wavelengths) between transmitter and receiver. Another problem is that carrier wave has to be watched continuously. Failures in carrier signal leads to errors in positioning. In addition, each cell Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

82

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

phone would need more signal strength to smooth such errors. This could lead to decrease of battery runtime.

Figure 6.4: Schematic overview of different GSM positioning architecture.

83

6.6 LOGICAL CHANNEL AND MAPPING The primary goal of any communication system is to transmit the user information. In cellular networks this means the transmission of traffic channels (speech or data) between MSs and the network. Reliable transmission of traffic channels requires also transmission of signaling on variety of logical control channels. However, all logical channels need to be mapped into some time slot at some carrier frequency for actual transmission. “Logical channels just describe specific usage of the physical resource”.

where

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

The full rate traffic channels carry information (that includes the error control coding) at the bit rate of 22.8 kbits/s. And the half rate traffic channels carry information at the bit rate of 11.4 kbits/s. In order to transmit signaling in parallel to the user data, which occupies a traffic channel, GSM offers two possibilities: Dedicated Associated Channels 1. A traffic channel (TCH) is always allocated together with a signaling channel known as slow-rate associated control channel - SACCH.  When the TCH is mapped to frames and multiframes, within a multiframe of 26 TDMA frames, only 24 TDMA frames carry TCH frames.  The 13th TDMA frame of each multiframe is reserved for SACCH.  This organization gives SACCH capacity of about 2 messages per second. SACCH can thus be used only for non-urgent signaling such as measurements of the received signal quality in the uplink and settings for power level and timing advance in the downlink. 2. For the fast associated signaling needs, the capacity of TCH can temporarily be stolen for signaling. Thus, the Fast Associated Control Channel (FACCH) means just a special usage of a TCH. FACCH is especially used for urgent handover signaling. “The fast and slow associated control channels are defined separately for the cases when the TCH is a full rate channel, and when it is a half rate channel. So we have FACCH/F and FACCH/H as well as SACCH/TF and SACCH/TH”. Dedicated Standalone Channels There is actually one more traffic channel called the TCH/8.  TCH/8 is a channel with exactly one eighth of the gross bit capacity of TCH/F.  The TCH/8s are never used alone, but they are always grouped to a group of 8 TCH/8s which has the gross bit capacity of TCH/F and is called SDCCH/8 or to a group of 4 TCH/8s, which has the gross bit capacity of TCH/H and is called SDCCH/4.  The SDCCHs are used mainly during call setup for registration, location updating, authentication, etc. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

84

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



The SDCCH/8 and SDCCH/4 have their own slow associated control channels, SACCH/C8 and SACCH/C4.

“The use of SDCCH improves the system efficiency, because the call setup signaling for 8 calls can be administered within the capacity of a single TCH/F”. Common Control Channels There are three types of common control channels. 

The paging channel (PCH) contains paging commands from the BTS to the MSs in case of an incoming call. Naturally this channel is only used for downlink.  The random access channel (RACH) on the other hand is only used in the uplink, when MSs are requesting the allocation of a free channel for communication.  The corresponding downlink channel is the access grant channel (AGCH) through which the BTS signals to a MS the allocation of a channel that MSs had requested via the RACH. Even though the PCH, RACH and AGCH are common resources for communication with all MSs, they normally are used to carry signaling that is specific to a single MS. Broadcast Channels The broadcast channels BCCH, FCCH and SCH are used in downlink only for signaling that is intended for all MSs. They all share the time slot 0 on one of the frequency channels of the cell.  The Broadcast Control Channel (BCCH) carries general information such as location area identity, the identity of the operator, frequencies of neighboring cells, and access parameters.  The function of the synchronization channel (SCH) is to enable frame synchronization of MSs and provide the identification of the BTS.  The Frequency Correction Channel (FCCH) is a special channel having a long sequence of zero symbols. It creates after modulation a sinusoidal carrier that is slightly above the center of the frequency channel by a fixed frequency offset. This sinusoidal signal can be used for carrier synchronization at the receiver. “In every cell one FCCH and one SCH must be broadcast, and in addition, every cell must have at least one BCCH, AGCH and RACH to support the access from the MSs”. The total available bandwidth of the GSM system is 2 x 25 MHz. Since the outermost carrier frequencies are located 200 kHz away from the band edges, this bandwidth corresponds to 124 frequency channels with 200 kHz carrier spacing. Each individual frequency channel is divided among 8 users using TDMA scheme. Thus, the basic physical resource for transmission of data in the GSM system is one of 8 timeslots of a TDMA frame.  The duration of a time slot is approximately 0.577 ms.  Uplink and downlink bursts of a mobile are separated in time by three time slots, so that the mobile station does not have to transmit and receive simultaneously which simplifies the MS hardware. The duration of a 26 multiframe, which consists of 26TDMA frames is exactly 120 ms.  Thus, the exact duration of a time slot is 120 / (26 x 8) =15/26 ms.  The cycle of 120 ms is needed for transmission of slow-rate signaling in SACCH in parallel to traffic channels. The SACCH is mapped to the 13th TDMA frame of the 26 multiframe instead of a traffic channel. Similarly, the duration of a 51 multiframe is approximately 51 x 8 x 0.577 ms = 235.385 ms.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

85

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



This cycle time is needed for transmission of slow-rate signaling in SACCH/C8 or SACCH/C4 in parallel to SDCCH/8 or SDCCH/4.

The superframe consists of 51 x 26 = 1326 TDMA frames which corresponds to either 51 X 26 multiframes" or 26 X 51 multiframes".  Its duration is 6.12 seconds which is the smallest cycle for which the organization of all channels is repeated, irrespective of whether "26 multiframes" or "51 multiframes" are used. The hyperframe consists of 2048 x 51 x 26 = 2715 648 TDMA frames.  Its duration is 12 533.76 seconds or about three and a half hours, which is a cycle required for ciphering. The hierarchy of TDMA frames is given below in Figure 6.7.

86

Figure 6.7: hierarchy of TDMA frames. Time slot represents just the allocation of time interval for transmission, the contents of time slots are called bursts.  The most important burst is the normal burst that is designed to accommodate one TCH/F.  The gross bit rate of TCH/F is 22.8kbit/s.  Within the 120 ms duration of a "26 multiframe" there are 24 TDMA frames available for traffic channels.  Thus, there are 24/(120/20) = 4 time slots (since a user can access only one time slot per time frame)available for each individual user within 20 ms, which is the basic speech coding frame.  As a consequence there are 20 ms x 22.8 kbit/s / 4 = 114 information bits to be transmitted in each normal burst.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

87

  

However, the information carrying capacity of the normal burst is 116 bits instead of 114 bits. The two extra bits (the bits closest to the training sequence) are the stealing flags. These bits indicate to the receiver, whether or not received burst contains TCH or FACCH signaling. The 26-bit training sequence in a normal burst is used for frame synchronization and equalizer training in the receiver.

“The training sequence is located in the middle of the burst in order to minimize the maximum distance between the training sequence and data bits, i.e. in order to minimize the possible error in the channel estimate computed from the training sequence when compared to the true state of the channel during data transmission”.   

The slight disadvantage of this middle location is that the first half of the burst must be stored in memory before it can be equalized, but this is not a major problem in practical implementation. Instead of using a single fixed training sequence, 8 different training sequences are used in normal bursts. When different training sequences are allocated for geographically adjacent users of the same frequency channel, the effect of cochannel interference in symbol synchronization and equalizer training can be minimized. “This naturally requires that the training sequences are chosen to have very low cross-correlation between each other”.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

The synchronization burst is used only in the downlink to carry SCH logical channel.  It has exactly the same length as the normal burst, but its contents are different.  Since the synchronization burst is normally the first bursts that a MS demodulates, longer training sequence of 64 bits is required to facilitate the initial symbol synchronization and initial equalizer training.  This training sequence is unique for necessity, since a MS could have no way of knowing which training sequence is used in the synchronization burst before receiving it. The frequency correction burst is used only in the downlink to carry FCCH logical channel.  The data part of this burst is just a long sequence of zero symbols, which creates after modulation a sinusoidal carrier, which can be used for carrier synchronization at the receiver. The access burst is used only in the uplink to carry RACH logical channel.  An extended training sequence of 41 bits is required in access burst since the situation is a lot more difficult to the receiver than the receiving of normal burst.  The access burst is the first transmission of a MS in a cell, and the receiver of the BTS cannot know beforehand the expected level of received signal, frequency error or exact timing within a time slot.  The MSs of GSM system are allowed to operate in a cell when their distance from the BTS is below 35 km.  The maximum time required for the radio wave to travel from BTS to MS and then back from MS to BTS is thus 233.33 μs.  Since the MSs lock their timing to the timing of the received frames from BTS, this time is exactly the duration of the worst case overlap between transmissions in successive uplink TDMA timeslots.  To prevent overlap between access bursts, a guard time greater than 234 μs is required. The guard time that is chosen in GSM standardization is 252 μs = 68.25 symbol durations.  A guard time greater than 234 μs takes over 40 % of the whole 577μs guard time duration of a time slot, which would be totally unacceptable in normal operation. The considerably shorter guard period of other bursts is possible since timing advance can be used in uplink transmission of other bursts.  By using the 41-bit training sequence of the access burst, the BTS can measure the delay of the received signal relative to a MS at zero distance.  This delay is transmitted to the MS, which advances the starting time of all subsequent transmissions by this amount of time with respect to the timing of the received frames from BTS.  Thus, when the uplink signal reaches the BTS it is nicely fitted within the correct time slot even with the reduced guard time of only 30.5 μs (8.25 bits). 6.7 GSM SPEECH CODING Full Rate Speech Coding The speech coding algorithm for TCH/FS channel of GSM is the Regular Pulse Excitation with Long Term Prediction linear predictive coder (RPELTP). It is a hybrid coder based on linear predictive coding with separate pitch filter (LTP) and improved coding for the excitation signal (RPE). The bit rate of the codec is 260 bit / 20 ms = 13 kbit/s, where 20ms is the duration of the basic analysis frame of the coder. The transmitted parameters for every 20 ms speech frame are:

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

88

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

    

  

The bits within 260-bit block are divided into three classes according to significance to the subjective quality of the output speech. There are 50 bits in the most important class Ia, 132 bits in the class Ib and 78 bits in the least important class II. For class Ia bits error detection by using a (53, 50) block code is used (i.e. 3-bit cyclic redundancy check (CRC) for error detection are added). After that the resulting 53 class Ia bits are combined with 132 class Ib bits and 4 tail bits (for Viterbi decoder of the convolutional code) for rate 1/2 convolutional encoding. Thus, the output of the convolutional encoder contains 2 x (50 + 3 + 132 + 4) = 378 bits. The class II bits are transmitted as they are without any error protection, and the total length of the output block from channel coding is 378 + 78 = 456 bits, which corresponds to bit rate 456 bit/ 20 ms = 22.8 kbit/s. The block of 456 bits is divided into eight 57-bit sub blocks for interleaving. After interleaving we have four 114-bit blocks, each of which can be assembled to the data part of a single normal burst together with two stealing flag bits. (However,) before burst assembling, each 114-bit block is encrypted to provide confidentiality to user data.

The following figure shows the block diagram of the transmission of TCH/FS logical channel.

Figure 6.8: block diagram of the transmission of TCH/FS logical channel. After RF processing, the first operation in the receiver is the equalization of the received signal. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

89

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

   

After this, the receiver executes exactly the inverse operations to the receiver: burst disassembly, decryption, deinterleaving, convolutional decoding, error detection, and speech decoding. The convolutional code is used to correct as many bit errors for class I bits as possible. If the 3 CRC bits still indicate errors within the class Ia bits after convolutional decoding, the speech frame is replaced by the previous valid speech frame, which is attenuated for the output. In the case of multiple successive erroneous speech frames, the output is gradually attenuated more and more to avoid annoying sounds at the output.

If a call is continued outside the acceptable coverage area without handover, the output of the speech decoder will contain relatively large amount of disturbance due to undetected erroneous speech frames. The block diagram of the transmission of TCH/FS logical channel includes two special blocks: the Voice Activity Detection (VAD) and the comfort noise generation. The purpose of these blocks is to improve the GSM system efficiency by decreasing interference level through discontinuous transmission of speech.  When the speaker is not active, the VAD informs the burst assembly and the RF transmitter, and the normal transmission is replaced by pause in transmission broken periodically (every 480 ms) by the transmission of special bursts that inform the receiver about voice inactivity condition.  In order not to produce annoying disturbances to a human listener when transfer of speech is switched on and off, the receiver must replace the speech output by "comfort noise" during speech inactivity periods. When we take into account the fact that a burst may also have stolen for FACCH signaling, which is indicated by the stealing flags, the complete rules for the speech decoding are as follows: if ((stealing flags indicate FACCH signaling) or (CRC indicates errors within class Ia bits)) substitute attenuated previous valid speech frame for output

90

else if (pause in discontinuous transmission) generate comfort noise for output else decode the current speech frame for output end In internetworking to other networks, the MSC is connected to PSTN by using standard PCM techniques. Thus, in the network side the input/output speech is not an analog signal but a digital PCM signal. Correspondingly the encoding/decoding process is actually a transcoding from or to digital 64 kbit/s PCM.  In most network implementations the transcoding takes place in the BTS, but the GSM standards also allow the transcoding to take place in the BSC or in the MSC. It can be considered how the capacity of TCH/FS channel is actually used.  

The modulation bit rate of GSM is 270.833 kbit/s per 8-TDMA channel. Thus, the effective bit rate per user is approximately 33.85 kbit/s which consists of the following components:

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

“Thus, the bit rate of the speech codec 13 kbit/s is less than 40 % of the total required bit rate of 33.85 kbit/s”. In general, a speech codec for mobile communications should simultaneously meet the objectives of high speech quality and low bit rate, even though these objectives are inconsistent. To solve this dilemma, GSM speech coding has evolved separately into two different directions:  half-rate coding for enhanced capacity and  enhanced full-rate coding for improved speech quality. 91

Half-Rate Speech Coding The half-rate speech coding algorithm GSM is the Vector Sum Excited Linear Prediction (VSELP) coder. It is a hybrid coder based on code-excited linear prediction (CELP) coding, where four different excitation modes with separated codebooks are used for modeling the excitation signal.  The half rate codec can provide speech quality that is near the quality of the full-rate RPELTPcodec.  The transmitted parameters for every 20 ms VSELP speech frame are:



Hence, the bit rate at the output of the codec is 112 bit / 20 ms = 5.6 kbit/s.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

    

These 112 bits are divided into 95 more perceptually important class I bits and 17 less important class II bits. A 3-bit CRC is computed for the 22 most sensitive class I bits. After the 3 CRC bits, the 95 class I bits and 6 tail bits are encoded with a rate 1/3 convolutional code. Thus, the output of the convolutional encoder contains 3 x (95 + 3 + 6) = 312 bits. However, 101 coded bits corresponding to the class I bits and tail bits are punctured, so that only 211 bits remain. The class II bits are transmitted without any error protection, and the total length of the output block from channel coding is 211 + 17 = 228 bits that corresponds to bit rate 228 bit / 20 ms = 11.4 kbit/s, which is exactly the capacity of the half-rate traffic channel TCH/HS.

Enhanced Full-Rate Speech Coding The enhanced full-rate (EFR) speech coding algorithm GSM is the Algebraic Code-Excited Linear Prediction (ACELP) coder, which is also a hybrid coder based on CELP coding. The excitation vectors in the ACELP codebook are algebraic codes, which has several advantages: no storage is required (since excitation vector scan be computed during the codebook search), excitation vectors are robust against transmission error, and most importantly the codebook can be searched very efficiently. The 20 ms speech frame is divided into four 5 ms subframes, and an optimum excitation vector is searched from the codebook separately for each subframe. The enhanced full-rate codec can provide speech quality that indistinguishable from 64 kbit/s PCM quality. The transmitted parameters for every 20 ms ACELP speech frame are: 92

Hence, the bit rate at the output of the codec is 244 bit / 20 ms =12.2 kbit/s, which is different from the 13 kbit/s rate of the standard RPE-LTP full-rate speech codec. A modified convolutional coding is used with EFR to accommodate it into the 22.8 kbit/s TCH/FS channel: an 8-bit CRC and rate ½ convolutional coding are used to protect the 65 most sensitive bits; 117 moderately important bits are protected by convolutional coding only; and the remaining bits are transmitted without any error protection. The EFR speech codec has been designed specifically for improved speech quality, but an additional advantage of the improved coding algorithm is that it also has higher tolerance to interference than the RPE-LTP codec. This improved interference tolerance allows higher network loading without degrading the overall speech quality. 6.8 CHANNEL CODING AND INTERLEAVING Even sophisticated equalization and detection methods are not good enough to provide acceptable level of BER in transmission over fading mobile radio channel.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

“To decrease the BER to acceptable levels, channel coding must be used”. This improvement of BER can be naturally achieved only by increasing the bit rate in channel coding, and thus by decreasing the spectral efficiency of the system. Interleaving is the process of reordering data bits into a different order in the transmitter, so that in there ordering of the data bits in the receiver, the burst errors due to fades are distributed over a long span. In GSM, thus,  456 encoded bits of a 20 ms speech frame are divided into eight 57-bit subblocks, and these eight subblocks are spread over eight consecutive bursts.  The interleaving delay is approximately eight times the duration of a TDMA timeslot, i.e. 8 x 4.615 ≈ 37 ms.  The maximum allowable speech coding delay is 20 ms (i.e., the duration of the basic analysis frame of the speech encoder), and other delays in the baseband and RF processing can amount to over 20 ms.  Hence, the overall delay of a GSM transmitter can be 80 ms. 

The channel coding and interleaving schemes that are used for different logical channels is shown in the following table.

93

6.9 FREQUENCY HOPPING The principle of GSM frequency hopping (FH) is that two successive TDMA bursts are transmitted in different frequency channels. With the TDMA frame duration of 4.6 ms, this leads to hopping rate of 217hops/s (1/4.6 ms).  Slow frequency hopping was introduced into GSM system for two main reasons: to achieve frequency diversity and interferer diversity. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



The main idea of frequency diversity is that when the difference of two frequencies is greater than the coherence bandwidth of the radio channel, fading at these frequencies are uncorrelated. In practice a distance of about 1 MHz is enough to decorrelate the fading.

When a MS is moving at high speed, the time interval between two successive bursts 4.615 ms is greater than the coherence time of the channel, and the fading of two successive bursts are uncorrelated.  Thus, in high speed motion we do not need FH to achieve decorrelation of successive bursts (on the other hand it does no harm either).  However, for stationary or slowly moving MSs (such as pedestrians), the time separation of two successive bursts is not large enough to provide decorrelation of fading.  For low speed motion the use of FH is essential to provide decorrelation of fading for successive bursts, otherwise the transmission quality over the radio link would be seriously degraded in locations of deep fades.  Thus, when we are using FH, it is very likely that among few successive bursts only one is lost due to a deep fade. It is then the function of error correction codes and interleaving techniques to recover the lost data from the successfully received data of other bursts.  The error correction and interleaving processes of GSM should normally manage a loss of one burst in five without significant degradation in speech quality. Besides frequency diversity, the FH also provides interferer diversity. In areas of high cellular traffic density the capacity of a cellular system is limited by interference from the system itself.  Since power levels of different MSs are different and the usage of different frequency channels varies in the network, for static allocation of frequency channels the signal-to-interference ratios of different frequency channels vary significantly, and relatively large number of frequency channels within each cell are not available to traffic at all.  If we can spread the interference approximately evenly between different frequency channels by using FH, we can make the signal-to-interference ratios of different frequency channels more equal.  which mean that the number of available frequency channels within each cell can be increased. Even when FH is used the fixed duplex spacing of 45 MHz is retained, and the carrier frequency of an uplink frequency channel is found simply by adding 45 MHz to the carrier frequency of a downlink frequency channel. o A MS receives a single timeslot on a specific downlink frequency channel, o switches to the corresponding uplink channel (45 MHz away) for transmission of a single timeslot, o receives the next timeslot on a different downlink frequency channel, o switches again to the corresponding uplink channel (45 MHz away), etc. Each hopping sequence, i.e., a sequence of time slot numbers and frequency channels that a MS can use, may use up to 64 different frequency channels out of 124 frequency channels of the GSM system.  Within them successive frequency channels are chosen by using a pseudorandom algorithm to distribute transmissions evenly among frequencies. However, these algorithms have been designed to avoid collisions (simultaneous use of same frequency channel) inside one cell and between a numbers of adjacent cells. The frequency hopping algorithm is broadcast to all MSs on the BCCH. o All GSM MSs must have the frequency hopping capability. o However, frequency hopping is only used in the cells, which have bad multipath propagation conditions.  Under any circumstances, frequency hopping is not utilized in the frequency channel that is carrying the broadcast common control channel of the BS, since a new mobile in a cell must be able to locate this channel prior to knowing anything about the frequency hopping sequence. The following figure illustrates the effect of frequency hopping on encoded (class I) and uncoded (class II) bits of TCH/FS for a MS with a velocity 5 km/h in frequency-flat fading environment.  The curves "no FEC, FH" and "no FEC, no FH" cannot be distinguished from each other, since frequency hopping does not improve the average BER without coding. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

94

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



However, when forward error correction coding is used, frequency hopping can decrease BER by a factor of 10 for large Eb/N0 ratios.

95

Figure 6.9: Effect of frequency hopping on encoded and uncoded bits of TCH/FS (for a MS with a velocity 5 km/h in frequency-flat fading environment). 6.9 GSM SGNALLING PROCEDURE There are generally four mechanisms that exist in voice-oriented mobile wireless networks and that allow a mobile to establish and maintain a connection to the network. These are registration (initiates once one turns on the cellphone), call establishment (when a user initiates or receives a call), handover (helps the MS to get connected from one BS to another), and security (protects the user from fraud and eavesdropping). Registration When a MS is switched on or when a MS has changed an area, a registration procedure takes place. However, before this the MS must be "in tune" with the network, which requires frequency synchronization, time synchronization and acquisition of basic information about the network. Thus, the MS has to: 1. Locate the radio channel, where the broadcast channels FCCH, SCH and BCCH are transmitted by identifying the FCCH from the received spectrum. By locking to the frequency in the FCCH, the MS acquires frequency synchronization with the network. 2. After finding the FCCH, the MS can locate the SCH, which provides the time synchronization, the information about the current frame numbering, and the training sequence of this cell. 3. After acquiring frequency and time synchronization, the MS is able to read BCCH, which contains the parameters of the cell, and especially the access parameters. Only after this procedure the MS can transmit on an uplink channel without interfering to other traffic in the cell. The following table shows the signaling traffic between different entities of the GSM network during the subsequent registration which includes also the authentication of the MS. The MS uses the RACH for requesting channel, and receives on the AGCH the channel assignment to a SDCCH for the exchange of signaling traffic. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 6.4: Traffic flow in GSM registration procedure.

Call Origination Before a call can be originated, the MS must be switched on and registered into the system. The following table shows the signaling traffic between the MS and the BTS during the call origination.  The MS again uses the RACH for requesting channel and receives on the AGCH the channel assignment to a SDCCH for the exchange of signaling traffic.  After the required signaling is finalized on the SDCCH, a traffic channel is assigned to the MS for the genuine speech traffic. Table 6.5: traffic between the MS and the BTS during the call origination.

96

`The network may tell the mobile to start ciphering its data. However, it is not mandatory for the network to enable ciphering, and the MS (or the subscriber) has no privilege to request for ciphering. The procedure for receipt of incoming call is almost identical to the call origination procedure. Paging On a given time instant, the MS may be anywhere in the operator's service area, or even in the service area of any operator. For the network to be able to route an incoming call to the subscriber, it has to localize Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

the MS. Thus, the radio network must be able to perform subscriber mobility management. One extreme way is to update localization information in the MSC as soon as the MS is locked on the signaling channel of a new BTS, i.e. to localize a MS with a precision of a single cell. But, this would require transmission of very large amount of signaling information in the network due to frequent location updates. The other extreme way is to localize a MS only with a precision of a GSM network and to send a paging message about an incoming call to a certain MS to all BTSs of the network. This method would also generate large amounts of signaling due to duplication of paging messages which would consume large part of the network capacity. In the GSM, a compromise between these two methods is used.  The BSCs and their BTSs are divided into groups called location areas and only the information about in which location area the MS is localized is updated in the MSC.  For routing of an incoming call, a paging message needs to be transmitted only through the BTSs in the associated location area. “Adding several BTSs to the groups will lower the amount of transmission in location updating, but increase the transmission of paging messages.” By a proper choice of the group structure, the total signaling traffic can be minimized. The location updating procedures, and subsequent paging, use the two location registers – HLR and VLR –of the Switching Subsystem.  When a mobile station is switched on in a new location area, or it moves to a new location area of different operator, it must register with the network to indicate its current location.  In the normal case, a location update message is sent to the VLR of the new MSC which records the location area information, and then sends the location information to the subscriber's HLR.  If the subscriber is entitled to service, the HLR sends a subset of the subscriber information needed for call control to the new MSC/VLR and sends a message to the old MSC/VLR to cancel the old registration. Handover GSM employs a technique called the mobile assisted handover (MAHO). In this technique, the MS must continuously monitor the power levels and error rates of the neighboring BTSs on the list that the MS has received from its current BTS. This monitoring is possible, since in a TDMA system the receiver of a MS is required for receiving signal from the serving base station only during a single time slot of the 8-slot TDMA frame. The measurement results of the power levels and error rates of the serving cell and the best six neighboring cells are periodically (at least once per second) sent to the serving BTS over the SACCH. The serving BTS and the neighboring BTSs may be simultaneously performing measurements on the quality and the power level of the MS. It is up to the network (operator) to act upon the quality or power measurements, and the handover thresholds can be adjusted according to changing environment and traffic load of different BTSs. The GSM system distinguishes between different types of handovers depending on whether a call is transferred between:  channels (time slots) in the same cell,  cells (BTSs) under the control of the same BSC,  cells under the control of different BSCs but belonging to the same MSC or  cells under the control of different MSCs. If both the old cell and the new cell are controlled by the same BSC, the handover can be handled by the BSC without assistance from the MSC, which anyhow has to be informed about the handover. If a MS is entering a cell controlled by a different BSC, then the MSC must control the handover. When a MS is entering a cell controlled by a different MSC, there is an extra complication that after managing the handover, the old MSC must transfer the call to the new MSC. From technical point of view, the handover between different MSCs may well take place between different operators and even between different countries. It is up to the commercial contracts between operators, whether these handovers are allowed or not. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

97

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Also intracell handovers from one channel to another within the same cell are possible in the GSM system. These kinds of handovers are necessary for shifting from a channel of high level of cochannel interference to a channel where the level of interference is not as large. As in the case of two different cells under the control of the same BSC, the intracell handover can be handled by the BSC without assistance from the MSC. Handover signaling takes place at the Radio Resource Management layer of the GSM protocol stack. Algorithms for decision, when a handover should be taken, are not specified in the GSM Recommendations. Manufacturers and operators have freedom to choose any appropriate algorithm for handover decision. Security The TDMA nature of GSM combined with frequency hopping facility (a user is scheduled in a timeslot in a total of eight timeslots-time frame within certain frequency range-a channel bandwidth) makes it much more difficult for an eavesdropper to lock into a desired user's signal than in the first generation analog systems. Still, GSM includes elaborate mechanisms to guarantee subscriber security over the radio interface. In the GSM system, user-related information is transferred besides data channels also over signaling channels which can contain the called and calling telephone number, short messages etc. To protect network operators and subscribers against security threats, the implementation of the following security features is mandatory in any GSM network on both the network side and mobile side:  subscriber identity confidentiality,  subscriber identity authentication,  user data confidentiality, and  signaling information confidentiality. From a user point of view it is not relevant whether the user-related data is contained in a traffic channel or a signaling channel. Therefore, the GSM system basically provides three security services:  authentication of the subscriber (for the affirmation of the identity of the user)  enciphering of the user-related data (for the user data confidentiality)  temporary identities (for the confidentiality of the user identity) One security service that is not provided by the GSM system is the authentication of the network (BTS) by the mobile. The GSM security issues center around the removable subscriber identity module (SIM), which every GSM subscriber receives from the operator. The fact “That subscriber information and many security features are contained within the SIM instead of the mobile allows secure GSM access to users through any GSM mobile”. When a subscriber switches on the mobile, he must enter a personal identification number (PIN) to identify himself as the owner of the SIM. Until the PIN is verified, the mobile cannot access the data in the SIM. However, most operators allow users to disable this function after the first registration, so that the PIN is no more required, when the mobile is subsequently switched on. To achieve the tasks related to authentication and ciphering key management, the SIM must contain:  a microprocessor for algorithm execution,  masked programmed ROM for operating systems as well as A3 and A8 algorithms,  RAM for algorithm execution and input/output buffering, and  nonvolatile EEPROM (electrically erasable programmable read-only memory)to store updatable data such as the authentication key and the temporary mobile subscriber identity. It is important to note that all the security mechanisms of GSM are under sole control of the operators: the users have no possibility to affect whether authentication, encryption, etc. are applied or not. Moreover, the users are not necessarily aware of what security features are in use. Authentication The SIM of any subscriber contains, among other parameters, the individual subscriber authentication key Ki and the authentication algorithm A3. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

98

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

   

When a mobile tries to access the network, it receives from the base station a random number R (0 to 2128–1). Next, mobile uses the A3 algorithm to compute a signed response S from R by using his own authentication key Ki. The response S is sent back to network, where it is compared to independently computed version of S to authorize or deny access. Prior to this procedure the network has to know the identity of the subscriber, since the authentication key Ki of this specific user must be used also in the network side.

99

Figure 6.5: Schematic overview of GSM authentication mechanism. “The secrecy of Ki is the basis of GSM security functions”. Ki is stored inside SIM and it is completely protected from reading. It is only internally accessible within SIM for computation of S. Besides secrecy of Ki, for good level of security, it is required that A3 is a oneway function, which means that computation of S from R and Ki should be easy, whereas, the computation of Ki from R and S should be (practically) impossible. The algorithm A3 is not defined in the GSM standard. Actually, operators have a complete freedom to choose A3 and Ki. The only requirement for compatibility is that R contains 128 bits and the corresponding response S must contain 32 bits. Since the SIM of an individual user is always provided by a specific operator, this operator-dependent authentication still allows complete inter-operator roaming. Authentication and charging are always performed by the operator of the home network; therefore there is no need for a standardized authentication algorithm common to all networks: every operator can use his own proprietary algorithm. Only the interface parameters have to be specified system wide. The data needed for checking the authenticity of a user is usually generated by the home network. The actual verification and thus the access control to the system is handled locally by the VLR, where the user is temporarily registered. The VLR obtains pairs of random numbers R and corresponding signed responses S from the AuC upon request by using the international mobile subscriber identity(IMSI) as the identity of the subscriber to be authenticated. As the VLR and the AuC may be thousands of kilometers apart, the VLR may store five authentication pairs (R, S) for each subscriber. Each authentication pair is used only once and must be discarded after being used. When the user has moved to another VLR, the new VLR requests the IMSI from the previous VLR. The previous VLR transfers all the unused authentication pairs along the IMSI to the new VLR. The use of public-key cryptography would allow the local verification of the response in the VLR without any secret information having to be transferred from AuC to the VLR. The main reasons, why public-key solution was not chosen, are the computational complexity of algorithms and the amount of data to be transmitted over the radio path. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Encryption To protect the user data from eavesdropping at the radio connection GSM includes encryption functions. Since GSM is purely digital system, the ciphering method does not depend on the type of the transmitted data, “Ciphering is applied to the whole information part of a normal burst (except the two stealing flags) irrespective of the contents”. As in all existing cellular mobile systems, encryption is used on the radio path only; end-to-end enciphering is outside the scope of GSM standards. The enciphering can only be activated after the identity of the user has been verified, i.e. after authentication. The practical enciphering and deciphering are carried out as stream ciphering by an exclusive or operation between 114 data bits of a normal burst and a pseudorandom bit sequence. The security of the system is based on the generation of this pseudorandom bit sequence from a secret key Kc, which is done by using a special algorithm called "A5". For compatibility, the A5 algorithm must be part of GSM standards. However, for increased security of the GSM system the original GSM encryption algorithm, which is known as A5/1 algorithm, is not published in the GSM standards, it is only distributed among manufactures and operators, who have signed a special contract. To export a GSM system using A5/1 algorithm outside Western Europe a license from a controlling organization is required. To make exporting easier, ETSI has developed a new simpler algorithm called A5/2, which is not under such a strict export control. Both algorithms can coexist in a network and special measures are employed to guarantee that a mobile capable to only A5/2 algorithm can access systems that are normally utilizing A5/1algorithm. Besides the key Kc, the A5 uses the TDMA frame number as its other input.

Figure 6.6: Schematic overview of GSM encryption procedure. For full-duplex communication, two different ciphering sequences must be generated in parallel for uplink and downlink, as shown in the following figure. The use of the 22-bit TDMA frame number as input of encryption algorithm guarantees that the period between periodic repetitions of the ciphering sequence (the hyper frame) is almost three and a half hours, which greatly exceeds the duration of a normal call. Key Distribution The mobile and the network must agree on the encryption key Kc before data transmission can start. In GSM system a mobile acquires the key during the authentication stage in a parallel process, and Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

100

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

subsequently stores it into non-volatile memory of SIM, so that it is available even after switching the mobile power off and on. The method to deliver the encryption key Kc to a mobile without exposing it to eavesdroppers in the radio interface is based on the same random number R, which is sent by the base station to mobile for authentication purposes. The computation of the encryption key Kc is exactly analogous to the computation of the signed response S from R by using algorithm A3 with the authentication key Ki, only difference is that a different algorithm, called A8, is used for computation of Kc. Same computation can be carried out in the network side also since A8, R and Ki are known in the network, thus there is no need to ever transmit Kc over the radio interface. As A3, A8 is not defined in the GSM standard, and operators have a complete freedom to choose A8 as well. As A3 also A8 is not implemented in the MS, instead it is stored in the SIM. Since the real-time encryption is done by using the standardized A5 algorithm, the operator-dependent key management with proprietary A8 algorithm does not endanger inter-operator roaming. Since A3 and A8 algorithms are always used together with common inputs, they are in most cases implemented as a singleA3/A8 algorithm that produces 96-bit output, 32 bits for S and 64bits for Kc. After the encryption key Kc is delivered to a mobile, the network issues a command to the mobile to start ciphering. From this time onward, all data from mobile is ciphered, including the acknowledgment to the ciphering command. At handover the necessary encryption information is sent from the old base station to the new one, the encryption key Kc remains unchanged. User Identity Protection Because the encryption is possible only after the authentication has been completed, there are some messages that cannot be encrypted. These include paging response, location update request, and connection management request. Since all the signaling up to and including the first message carrying a nonambiguous subscriber identity must be sent in clear, a third party could at this stage listen to this identity, and know where a particular subscriber roams at this particular moment. The fact that a subscriber is located at a specific cell and is currently involved in a telephone call is critical information that should not be exposed to third parties. Thus, the permanently assigned international mobile subscriber identity (IMSI), which is contained in the SIM, should not be transmitted as such over radio interface. Instead a temporary alias, the temporary mobile subscriber identity (TMSI), that is valid for a specific location area, is assigned to all subscribers. “The IMSI identifies the country and network as well as personal identification number of a subscriber, whereas the TMSI has random content. Within a specific location area a TMSI uniquely corresponds to a single IMSI, but a third party cannot map TMSIs to IMSIs”. The TMSI must naturally be agreed by the mobile and the network by using protected enciphered signaling procedures. The TMSI is assigned during the location updating procedure, and is used as long as a subscriber remains active in the network. The mobile uses the TMSI when it reports to the network or originates a call. Similarly, the network uses the TMSI to page a mobile. The assignment, administration, and updating of the TMSI are performed by the VLR. When a mobile is switched off, it stores its TMSI on the SIM to ensure that it is available when it is switched on again. However, the IMSI has to be used for the setup of a session if there are no other means to identify a subscriber. This happens for instance, when the subscriber uses his SIM for the first time or at the data loss in the VLR, where the subscriber is temporarily registered. When the SIM is used for the first time, the MS will read the default TMSI, which is stored in the SIM, and send this value to the network (to the VLR). As this TMSI is unknown to the VLR, it will request the IMSI from the mobile. It assigns a TMSI to the subscriber and transmits this identifier in an enciphered form (after successful authentication) to the mobile. The mobile deciphers the TMSI and stores it and the location area identification (LAI) to the SIM. From then on this TMSI will be used by the mobile instead of the IMSI until a new TMSI has been assigned to the subscriber. A new TMSI is assigned at each location update. If there is no malfunction of the system, the IMSI will never again be used for call setup. Even if the SIM has moved to a new VLR in a different network, the Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

101

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

new VLR can obtain the IMSI from the previous VLR by using the old TMSI and LAI which have been sent by the mobile. The LAI identifies the country and operator of the previous network, so that it is internationally distinguishable. It should be emphasized that cryptographic security mechanisms –authentication, encryption, key distribution and user identity protection – form only a small part of the complete security framework of GSM or any other cellular network. The comprehensive security of the systems includes also topics like physical protection of equipment rooms, casing for apparatus as well as operating procedures of operating staff. 6.10 GSM TRANSMISSION The bursty transmission of MSs has a distinctive effect on the spectrum of the transmitted signal, and thus affects significantly the adjacent channel interference. Consider a case where the transmitter power is just abruptly switched on and off at the beginning and end of a burst, we can express the resulting signal xT(t) as multiplication of a continuous modulated signal xc(t) and a train of rectangular pulses.

102

In GSM system τ is the duration of a timeslot 0.577 ms and T0 is the duration of TDMA frame 4.615 ms. In frequency domain, the instantaneous switching of power corresponds to convoluting the Fourier transform of the continuous modulated signal xc(t) with a train of impulses, the weights of which can be calculated from a sinc function.

    

The width of the main lobe of sinc function is 2/τ, so that the convolution spreads the spectrum of the modulated signal approximately by 2 / 0.577 ms= 3.5 kHz. But besides the main lobe, the spectrum of the rectangular pulse train (the sinc function) contains significant components up to very high frequencies. These high frequency components can be attenuated, if smooth ramping up and down of the transmitter power is used instead of instantaneous switching on and off. One ramping function that has favorable rapidly attenuating spectrum is the raised cosine function. The ramping up and down cannot be done during the information carrying part of any burst. However, since all bursts contain the guard period, we have 8.25 symbol durations or 30.5 μs for non-instantaneous ramping up and down.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 6.8: Magnitude transfer function of a raised cosine filter. 103

Figure 6.9: Pulse Shape with a Raised Cosine Spectrum. To reduce cochannel interference in the GSM network and to reduce the power consumption of a mobile set, all MSs are required to support adaptive RF power control.  The minimum transmission power level for all MSs is 13 dBm (20 mW).  There are five different power categories of MSs according to the maximum power level of 29, 33, 37, 39 or 43 dBm (0.8, 2, 5, 8 or 20 W).  Between the minimum and maximum power level there is 2 dB spacing between adjacent power levels.  The initial power level for transmission of the first access burst from any MS is broadcast by the BTS on the BCCH. During transmission from MS, the BTS measures the level and quality of received signal and calculates from this data for the MS the minimum transmission level that provides acceptable transmission quality.  Power control is totally managed by the BTS. It sets the transmission level of a MS by issuing a power level command on downlink.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Changes in power level are restricted to take place at 60 ms time intervals. Except at the beginning of channel use, a power control command from BTS to MS does not trigger an immediate change in the MS transmitter power. “The maximum rate of change is restricted to be 2 dB for each time interval of 60 ms”. The following figure illustrates the changes in transmitter power of a MS that receives successive power control commands of 17, 37 and 35dBm.

104

6.11 GSM RECEIVER The GSM standards specify precisely the implementation of a GSM transmitter: multiplexing, channel coding, modulation, RF characteristics like spectrum, phase accuracy, intermodulation, frequency tolerance etc. This is absolutely necessary, since a bad transmitter can cause significant interference to other GSM users and also to users of other radio systems. However, with the exception of the speech decoder most parts of the receiver are not precisely specified. For example, GSM standards do not specify, whether a receiver should include an equalizer or not. The only thing that is defined is the overall receiver performance in terms of BER (or frame erasure rate FER or residual bit error rate RBER) at a specified reference sensitivity level (i.e. at a specified SNR) and at a specified reference (cochannel) interference level. From these specifications, the manufacturers of GSM receivers have to figure out what kind of detection, equalization, channel decoding etc. to use.  For testing a receiver the reference sensitivity level and the reference interference level are generated by using the GSM computational channel models at different velocities of the MS.  The TU3, TU50, RA250 and HT100 conditions are used where the number after the name of channel model indicates the velocity of the MS in km/h.  The performance has to be met at the reference sensitivity level that is -100 dBm for a DCS1800 mobile station, -102 dBm for a handheld GSM900 MS, and -104 dBm for otherGSM900 MSs.  For example, for class II bits of the full-rate speech channel, the ratio of the number of bits with undetected errors to the total number of transmitted bits in TU50 conditions is required to be below 8 %. Since the class II bits of TCH/FS are transmitted without any channel coding, this requirement effectively defines the performance level of detection and equalization in a GSM receiver.  When we consider mobile speeds up to 250 km/h at the carrier frequency of 900 MHz, we have to consider Doppler shifts up to 200 Hz, or coherence times of the radio channel as low as 0.8 ms.  However, the duration of a time slot in GSM is only 0.6 ms, and the channel can be approximated to be constant during one timeslot even at the extreme conditions. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



The consequence is that the channel estimate that is calculated from the training sequence of a normal burst can be used for equalization of the complete burst.  However, since the duration of a TDMA frame is 4.6 ms, it is obvious that there can be significant variations in the channel characteristics between two successive time slots of the same user.  Thus, a channel estimate that is calculated from the training sequence, one normal burst is not valid for the subsequent burst. Instead, successive bursts should be equalized independently from each other.  Situation for the American TDMA system with 6 timeslots in a frame and 48.6 kbit/s symbol rate is quite different. Here the duration of a time slot is 6.6 ms, and there can be significant variations in the channel characteristics within a single time slot.  Thus, an effective equalizer for the American TDMA should be able to follow variations in the channel within the duration of a single time slot. These changes naturally cannot be estimated from the training sequence. Instead they must be estimated by using the detected data portion of a slot, which is less reliable. However, it should be kept in mind that with symbol duration of 20.6 μs, the level of ISI in the American TDMA system (even with large delay spread of the radio channel around 10 μs) is very low when compared to the level of ISI in GSM.

105

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 07 GENERAL PACKET RADIO SERVICE ABBREVIATIONS 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

ANSI American National Standards Institute APN access point name AuC authentication center BSCs base station controllers BSS base station subservice BTS base transceiver stations CDMA code division multiple access CLNS connectionless network service CONS connection oriented network service EDGE enhanced data rates for GSM evolution

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

EIR equipment identity register ETSI European Telecommunications Standards Institute GGSN gateway GPRS support node GMM/SM GPRS mobility management and session management GPRS General Packet radio service GSM global system for mobile communications GSN GPRS support node GTP GPRS tunneling protocol HLR home location register HSCSD high-speed circuit-switched data

21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

IMSI International Mobile Subscriber Id IMT 2000 International Mobile Telecommunications 2000 IP Internet protocol IPsec Internet protocol security LA location area LLC logical link control MAC medium access control MO mobile originated MS mobile station MT mobile terminal, mobile terminated

31. 32. 33. 34. 35. 36. 37. 38. 39. 40.

MSC message switching center NSS network switching subsystem NSAP network service access point NSAPI network service access point identifier PCU packet control unit PIN personal identification number PDN packet data network PDP packet data protocol PDU protocol data unit, packet data units PLMN public land mobile network

41. 42. 43. 44. 45. 46. 47. 48. 49.

PTP point-to-point PTM point-to-multipoint RA routeing area RF radio frequency RLC radio link control SAP service access point SGSN serving GPRS support node SNDCP subnetwork dependent convergence protocol TDMA time division multiple access

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

106

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

50. TE terminal equipment 51. 52. 53. 54. 55.

TID tunnel identifier TLLI temporary logical link identity UMTS universal mobile telecommunications systems VLR visitor location register WCDMA wideband CDMA 7.1 GPRS OVERVIEW

GSM offers data services already but they have been constrained by the use of circuit-switched data channels over the air interface allowing a maximum bit rate of 14.4 kbit/s. Existing GSM data services do not fulfill the needs of users (or operators). From the user's point of view, data rates are too slow and the connection setup takes too long. From the technical point of view, these drawbacks result from the fact that current wireless data services are based on circuit-switched radio transmission. “The General Packet Radio Service (GPRS), a data extension of the mobile telephony standard GSM is true packet-switched architecture to allow mobile subscribers to benefit from high-speed transmission rates and run data applications from their mobile terminals”. For this reason, the GSM standard has continued its natural evolution to accommodate the requirement for higher bit rates.  The high-speed circuit-switched data (HSCSD) are one solution that address this requirement by allocating more time slots per subscriber and thus better rates. It remains however insufficient for bursty data applications such as Web browsing.  Moreover, HSCSD rely on circuit switching techniques making it unattractive for subscribers who want to be charged based on the volume of the data traffic they actually use rather than on the duration of the connection.  Service providers need effective means to share the scarce radio resources between more subscribers.  In a circuit-switched mode, a channel is allocated to a single user for the duration of the connection. This exclusive access to radio resources is not necessary for data applications with the use of packet switched techniques.  GPRS stands out as one major development in the GSM standard those benefits from packet switched techniques to provide mobile subscribers with the much needed high bit rates for bursty data transmissions.  It is possible theoretically for GPRS subscribers to use several time slots (packet data channels) simultaneously reaching a bit rate of about 170 kbit/s.  Volume-based charging is possible because channels are allocated to users only when packets are to be sent or received.  Bursty data applications make it possible to balance more efficiently the network resources between users because the provider can use transmission gaps for other subscriber activities. GPRS has been standardized by ETSI as part of the GSM phase 2+ developments represents the first implementation of packet switching within GSM.  The implementation of GPRS brought Internet Protocol (IP) capability to the GSM network, which enables connection to a wide range of public and private data networks.  GPRS is ideal for bursty data applications such as email or Internet access, and can also enable "virtual permanent connection" to data sources.  GPRS has been originally developed for GSM, but it is also integrated to the North American IS136 TDMA system. In the initial release, GPRS uses the same modulation as GSM (GMSK). The subsequent evolution of packet-based services in GSM introduces EDGE technology.  GPRS introduces two new nodes for handling packet traffic: the serving GPRS support node (SGSN); and the gateway GPRS support node (GGSN). These nodes interwork with the home Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

107

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects





   

location register (HLR), the mobile switching center/visitor location register (MSC/VLR) and base station subsystems (BSS). User data-for example, from a GPRS terminal to the Internet-is sent encapsulated over the IP backbone. In the GPRS standard, three new types of mobile terminal have been defined: Class A terminal, which supports simultaneous circuit-switched and packet switched traffic; Class B terminal, which supports either circuit-switched or packet-switched traffic (simultaneous network attachment) but does not support both kinds of traffic simultaneously; and Class C terminal, which is attached either as a packet-switched or circuit-switched terminal. To support efficient multiplexing of packet traffic to and from mobile terminals, a new packet data channel (PDCH) has been defined for the air interface. One PDCH is mapped onto a single time slot, thereby utilizing the same physical channel structure as ordinary circuit-switched GSM channels. All radio resources are managed from the BSC, where the pool of physical channels for a given cell can be used as either circuit switched GSM channels or packet data channels. By means of packet multiplexing, the allocated PDCHs can be shared by every GPRS user in the cell. The number of PDCHs in a cell can be fixed or dynamically allocated to meet fluctuating traffic demands. Thus, physical channels not currently in use by the circuit-switched service can be made available to GPRS traffic (Figure 7.1). More than one time slot can be allocated to a user during packet transfer. Uplink and downlink resources to connections are allocated separately on a case-by-case basis, which reflects the asymmetric behavior of packet data communication. 108

Figure 7.1: In this example, one time slot is statically assigned to GPRS; all other time slots are defined as dynamic GPRS resources. Benefits of GPRS The general packet radio service (GPRS) emerged into the market to optimize the Internet/intranet access capabilities and to enable other new wireless data services in GSM. The main benefits of GPRS are:  

high variable bit rates from a few bits per second up to 171.2kbit/s; instant information delivery as the need arises without the overhead of establishing a circuitswitched connection ("GPRS users are always connected");

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

    

possibility for volume based charging, enabling the user to stay on-line while paying only for actually transferred data; applicability for existing services in fixed-line packet networks (including Internet/intranet access) as well as for new applications; spectrum efficiency due to packet switching in the radio interface so that GPRS radio resources are used only when users are actually sending or receiving data and improved the available radio resource can be shared between several users; improved sharing of network resources through use of packet-switched technology in the network infrastructure; Worldwide support, since the GPRS service is not only deployed on GSM networks, the North American IS-136 TDMA standard also supports GPRS.

Drawbacks of GPRS GPRS offers a major improvement in spectrum efficiency, capability and functionality compared with today's GSM data services. However, it is important to note that GPRS also has some drawbacks:    

shortage of cell capacity, since introduction of GPRS will increase traffic demand in network, when conventional GSM calls and GPRS calls both have to use the same resources; limited practical bit rates, since it is unlikely that a network operator would allow all eight timeslots to be used without any error protection by a single GPRS user to realize the maximum GPRS data transmission speed 171.2 kbit/s; problems with GPRS mobile terminated calls that may not be supported due to problems of charging of incoming IP traffic, which can hinder migration of applications from fixed-line networks; absence of store and forward mechanism, so that delivery of GPRS messages is guaranteed only during time periods when a MS can be reached by the network.

GPRS Services GPRS services are defined to fall in one of two categories: point-to-point (PTP) and point-to multipoint (PTM) services.  The PTM services provide the subscribers with the capability to send data to multiple destinations within one single service request.  Table 8.1 illustrates the general description of the PTP services and some possible applications. Table 8.2 shows a general description of these services and some possible applications.  With the exception of point-to-multipoint multicast (PTM-M) services, groups must be defined and members are required to join an ongoing call to become participants.  A point-to-multipoint group (PTM-G) call is usually restricted to members located within a specific geographical area.  An IP-multicast (IP-M) call is on the other hand independent of the geographical area of the participants and can be internal to the network or distributed across the internet.

Table 7.1: Point-to-point (PTP) GPRS services

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

109

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 7.2: Point-to-multipoint (PTM) GPRS services.

110

7.2 GPRS ARCHITECHTURE GPRS is considered as a service or feature of GSM. Designed by ETSI was implemented over the existing infrastructure of GSM without interfering with the already existing services. The aim is quick GPRS deployment with minor impact on existing GSM PLMN components. Figure 7.2 illustrates the logical architecture of a GSM network supporting GPRS.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 7.2: GPRS reference model. The GPRS terminals GPRS and GSM systems provide inter-working and sharing of resources dynamically between users. For this reason, three types of terminals have been defined:   

A class-A MS can carry a circuit switched and a packet switched connection simultaneously enabling the subscriber to initiate or receive a voice call without interrupting a data transmission or reception activity. An MS of class-B is able to connect to both GSM and GPRS at the same time but an incoming voice call requires GPRS data transactions in progress to be suspended for the duration of the call. GPRS data transactions can then resume at the end of the voice call. Finally, a class-C MS allows subscribers to access one service type only at a given time in an exclusive manner.

The GPRS MS has two components: mobile terminal (MT) which is typically a handset used to access the radio interface as a radio modem, and terminal equipment (TE) which is typically a laptop or a personal digital assistant (PDA). GPRS BSS GPRS has minor impact on the existing GSM BSS making it easy to reuse existing component and links without major modifications. This is possible because GPRS uses the same frequency bands and hopping techniques, the same TDMA frame structure, the same radio modulation and burst structure as GSM. A new functional component called packet control unit (PCU) was added to the BSS in the GPRS standard to support the handling of data packets. The PCU is placed logically between the BSS and the GPRS NSS. Unlike the voice circuit connections however, connections in GPRS have to be established and released between the BSS and the MS only when data need to be transported over the air interface. GPRS NSS The GPRS NSS can be viewed as an overlay network ensuring the link between mobile users Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

111

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

and data networks. GPRS introduces a new functional element to the GSM infrastructure GPRS support node (GSN) which can be either a serving-GSN (SGSN) or a gateway-GSN (GGSN). This addition is necessary for the GSM network in order to support packet data services.  The network is generally divided into several service areas controlled by separate SGSNs.  Only one SGSN serves an MS at a given time provided it is located in its service area.  The SGSN is primarily responsible for keeping track of the MSs it serves, and for access control to data services.  The GGSN on the other hand provides the interface to external packet data networks (PDNs).  The SGSN is connected to the BSS by frame relay and to possibly several GGSNs via a GPRS backbone network. The HLR database is updated to contain GPRS subscriber information. Adaptations to an existing MSC/VLR are not required but the GPRS standard suggests some enhancements to coordinate between the SGSN and the MSC/VLR if the optional interface between the two is to be supported.

112

Transmission/signaling planes in GPRS A layered protocol structure is adopted for the transmission and signaling planes in GPRS (Figure 7.3).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 7.3: GPRS transmission/signaling planes from MS to GGSN. 

The sub-network dependent convergence protocol (SNDCP) serves as a mapping of the characteristics of the underlying network such as IP.  Mobility management functionality is supported by the GPRS mobility management (GMM) and session management (SM) layers.  The logical link control (LLC) layer provides a logical link between the MS and the SGSN and manages reliable transmission while at the same time supporting point-to-point and point-tomultipoint addressing.  The radio link control (RLC), medium access control (MAC), and GSM RF (radio frequency) layers control the radio link, the allocation of physical channels and radio frequency.  LLC PDUs (packet data units) between the MS and the SGSN are relayed at the BSS.  The base station system GPRS protocol (BSSGP) layer handles routing and QoS between the BSS and the SGSN.  The GPRS tunneling protocol (GTP) is the basis for tunnel signaling and user PDUs between the SGSN and GGSN. The others units have similar functionalities as already described in GSM and are not described further. 7.3 PACKET-SWITCHED TRANSMISSION OVER THE AIR INTERFACE User data packets are segmented, coded and transformed into radio blocks.  Each radio block is further interleaved over four standard GSM normal bursts that is, over the same basic vehicle that carries coded, circuit-switched speech across the air interface.  When errors occur, data packets can be retransmitted at the radio block level. The set of  bursts that results from a single user data packet is marked with a temporary flow identifier (TFI), which is used on the receiving side to reassemble the user data packet.  A new set of logical channels has been defined for GPRS traffic. This set includes control channels and packet data traffic channels.  A physical channel allocated for GPRS traffic is called a packet data channel (PDCH). One or more physical channels in a cell can be statically or dynamically assigned for PDCHs. Static PDCHs are always available, whereas dynamic PDCHs are provided on a case-by-case basis.  The PDCH consists of a multiframe pattern that runs on time slots assigned to GPRS. This  is basically a predefined pattern of GPRS control channels and data traffic channels that keeps repeating itself.  In cells defined as having only dynamic GPRS resources and which only run circuit-switched channels, the GPRS terminals use the circuit-switched control channels until one or more PDCH are assigned. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

113

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

  

Several mobile terminals can dynamically share the pool of packet data channels in a cell, and several PDCHs can be used simultaneously for a single connection. Thus, a user data packet can be transmitted over multiple packet data channels and reassembled at the other end (Figure 7.4). The network side controls the allocation of resources.

To start packet transmission on the uplink,  the mobile terminal requests resources.  The network tells the terminal which PDCHs to use.  The network also sends a flag value which, when it occurs on the corresponding downlink, tells the mobile terminal to begin transmitting. To start packet transmission on the downlink,  the network sends an assignment message to the mobile terminal, indicating which PDCHs will be used and the value of the TFI assigned to the transfer.  The mobile terminal monitors the downlink PDCHs and identifies its packets via the TFI.

114

Figure 7.4: GPRS users share the pool of resources in a cell. 7.4 SECURITY IN GPRS GPRS provides a security function similar to that of GSM.  It is responsible for authentication and service request validation to prevent unauthorized service usage.  User confidentiality is also protected using temporary identification during connections to the GPRS network.  Finally, user data are protected using ciphering techniques. Authentication in GPRS Authentication in cellular systems often means the use of a PIN code as a means of identification. Such method is not very secure since it is possible in a radio environment to capture the PIN and therefore to break the confidentiality of the subscriber. Especially dangerous is the fact that this PIN is assigned once at subscription and therefore it can be captured in many ways. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The GSM/GPRS approach addresses this problem by varying the access code for every connection.  A secret parameter Ki specific to the user is used to compute a shared value using an operatordependent one-way (trap-door) algorithm. Figure 8.5 shows the authentication computation using the parameter Ki which is never transmitted via the air interface.  The sequence of the figure is triggered when the MS requests to attach to the GPRS network or may be requested by the SGSN if the MS is roaming.

Figure 7.5: The GPRS authentication computation   

The MS is issued a new random number RAND whenever authentication is required. The MS computes the value SRES (signed result) using Ki and RAND and forwards it back to the SGSN for comparison. The authentication procedure in GPRS is executed from the SGSN instead of the MSC/VLR as in GSM. However, the procedure is the same as in the case of GSM.

Ciphering in GPRS Ciphering between the MS and SGSN starts only after the MS is authenticated. Similarly,  A secret ciphering key Kc is used at the MS and the SGSN to encrypt the exchange of messages.  The ciphering scope is from the MS to the SGSN while in GSM it is from the MS to the BSS. This is done to simplify the key management since cell selection by the MS can occur frequently and therefore packets may travel via different BTSs.  In GSM however, a single logical channel is used between the MS and the BTS at each given time.  Also, by extending the scope of ciphering in GPRS, ciphering algorithms can be changed without updating every element of the BSS. Security in the GPRS backbone Authentication of subscribers in GPRS is performed by the SGSN, and the scope of ciphering is between the MS and the SGSN. The GPRS standard has left to the implementation the requirements for security beyond the SGSN and GGSN.  For instance, an IP-based GPRS backbone can ensure security using IP security protocol (IPsec) which groups a set of mechanisms used to protect traffic at the IP level.  Secured services offered by IPsec is integrity in connectionless mode, authentication of data origin, and protection of data confidentiality.  IPsec provides security of data at the network level rather than at the application or physical levels. An operator can ensure security in the GPRS backbone by relying on several options such as the following.  installing firewalls at the Gi reference point (between a GGSN and an external data network), Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

115

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

  

using border gateways when interfacing to other GPRS networks, improving the backbone network protection by using some security products such as IPsec for an IP-based backbone as mentioned above, and using some packet filtering mechanisms. 7.5 MOBILITY MANAGEMENT IN GPRS

Mobility management is the means by which a mobile network such as GPRS can keep track of the mobile subscriber location while connected to the network. GPRS service areas In GSM, the network is divided into several MSC/VLR service areas.  Each MSC/VLR spans over a group of location areas (LAs) which are sets of cells.  Figure 7.6 illustrates a simplified example of the GSM network service areas.  The network is shown to be divided into five LAs and two MSC/ VLR service areas.  The thick line in the figure is used to show the separation between the two service areas.

116

Figure7.6: GSM network service areas. In GPRS on the other hand, a group of cells is called a routeing area (RA).  The SGSN controls a service area containing several RAs.  There may not be a direct mapping between SGSN and MSC/VLR service areas but an RA is a subset of one, and only one, LA.  GPRS has chosen a different layout from GSM (i.e., RAs instead of LAs) to allow for signaling and paging over geographically smaller areas and thus, a better optimization of radio resources.  One possible implementation of GPRS in the existing GSM network of Figure 8.5 is shown in Figure 7.7.  The example suggests three SGSN service areas to span over 11 RAs. Note that in a real network implementation, the layout is decided by the operator of the network.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 7.7: GPRS network service areas. 117

Accessing the GPRS network An MS can connect to the GPRS network by requesting a GPRS attach procedure.  The outcome is the establishment of a logical link between the MS and a single SGSN and the creation of a mobility management context.  GPRS attach and PDP context activation must be executed in order for GPRS users to connect to external packet data networks.  The mobile terminal makes itself known to the network by means of GPRS attach-GPRS attach corresponds to IMSI attach, which is used for circuit-switched traffic.  Once the terminal is attached to the network, the network knows its location and capabilities. If the unit is a class A or class B terminal, then circuit switched IMSI attach can be performed at the same time (Figure 7.8).  The mobile terminal requests that it be attached to the network.  The terminal’s request, which is sent to the SGSN, indicates its multi-slot capabilities, the ciphering algorithms it supports, and whether it wants to attach to a packet-switched service, a circuit switched service, or to both.  Authentication is made between the terminal and the HLR.  Subscriber data from the HLR is inserted into the SGSN and the MSC/VLR.  The SGSN informs the terminal that it is attached to the network.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 7.8: GPRS attach. Mobility management states The MS in GSM can be in one of two states: Idle or Dedicated.  A channel allocation is held for the MS exclusively when it is in Dedicated mode due to the nature of circuit- switched connections.  When the connection is released, the MS returns to Idle mode.  A GPRS MS on the other hand can share radio channels with other subscribers connected to the network. For this reason, the MS is defined to have three possible states: Idle, Ready, and Standby (Figure 7.9).

Figure 7.9: GSM and GPRS functional state models. Idle state An MS in the Idle state is not traceable and can only receive PTM-M transmissions such as general broadcast events destined to a specific geographical area. The MS needs to perform the attach procedure in order to connect to the GPRS network and become reachable. Ready state Data are sent or received in this state.  The MS informs the SGSN when it changes cells.  The MS may explicitly request (or can be forced by the network) to detach in which case it moves to Idle.  A timer monitors the Ready state and upon its expiry, the MS is put on Standby. The timer insures that resources are not wasted by an inactive MS. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

118

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Standby state A connected MS which is inactive is put in the Standby state. Moving back to Ready can be triggered by sending data or signaling information from the MS to the SGSN.  Upon arrival of data destined to the MS, the SGSN pages the latter and a response to the page moves the MS back to the Ready state.  The MS may wish (or can be forced by the network) to terminate the connection by requesting to detach in which case it returns to Idle.  A timer is used by the SGSN to monitor the tracking of the MS, and when it expires, the MS is detached and is considered unreachable. Keeping track of the MS Location management is the means by which the GPRS network keeps track of the MS location. Within a GPRS network, three types of location management procedures are described (Figure 7.8):  Cell update is the means by which an MS informs the network of its current cell location.  Intra-SGSN routeing update is the procedure used when an MS changes RA and remains serviced by the same SGSN.  Inter-SGSN routeing update is the procedure used when the entry of an MS to a new RA triggers a change of SGSN service area. 119

Figure7. 8: GPRS location management procedures. When roaming agreements between different network operators exist,  an MS that enters a new network performs a routeing update procedure provided that this is allowed by the implementation.  Otherwise, the MS is forced to the Idle state. The location management procedures depend on the current state of the MS (Table 7.3).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 7.3: Location management and mobility management states.

 

  

      

An Idle MS does not perform any updates. An MS in Standby performs routeing area updates only and does not inform the SGSN of its cell changes. The SGSN needs to know the cell changes only when the MS is in the Ready state. This is done in two ways: o when the new cell is within the same RA, a cell update takes place. o Alternatively when the new cell is within a new RA, then a routeing update procedure is performed instead. The BSS usually adds the cell identifier to the routeing update request and this will be used by the SGSN to derive the new RA identity. Also, even if the MS does not change RA, it is requested that an RA update be done periodically. The MS detects that it has entered a new cell or new RA by listening periodically to special control channels that broadcast general information such as the identities of the cell, the RA, the LA, and of the network. From the MS point of view, inter and intra-SGSN updates are transparent and the request is the same. The SGSN on the other hand is able to detect whether the MS is new in its service area or if it is already a serviced MS switching RAs. The MS includes the parameter old RA in the routeing update request when it enters a new RA. By looking at this parameter, the SGSN can conclude whether the old RA is in its service area or not. An old RA outside the scope of the SGSN implies that the MS must have been served by another SGSN and an inter-SGSN routeing update procedure is required (Figure 7.9). The old SGSN provides the new SGSN with the context information which describes the current activities of the MS. The new SGSN is responsible for informing the GGSNs and the HLR (which could be in another service area) of the new MS location. Finally, the HLR needs to provide the subscriber’s information to the new SGSN.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

120

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 7.9: Inter-SGSN routeing update procedure (simplified). 7.6 END-TO-END PACKET ROUTING To use data services via the GPRS network, specific procedures for packet data networks access and data routing are performed. Once an MS is attached, it can request to activate one or more packet data protocol (PDP) contexts which specify the packet data networks (PDNs) it wants to access. In other words, the MS asks the SGSN to create routing paths or tunnels to the external data networks. A PDP context activation procedure is initiated for each required PDP session. The activation procedure can be triggered by the MS (MS initiated) or by an incoming request from a PDN (network requested). “A PDP context refers to the parameters required to transfer packets between the MS and the PDN via a GGSN”. These parameters are specific to each PDP context and include routing information and the quality of service (QoS) profile. PDP context activation procedure 

The MS specifies its network service access point and the access point name (APN) of the PDN it wants to connect to. The APN specifies the target PDN network identifier such as intranet.company-name.com and the operator domain name such as operatorname.country.gprs. The SGSN identifies the corresponding GGSN and makes it aware of the MS. A two-way point-to- point path or a tunnel, is uniquely identified by a tunnel identifier (TID) and is established between the SGSN and the GGSN.

  

“Tunneling is the means by which all encapsulated packets are transferred from the point of encapsulation to the point of decapsulation”.   

In this case, the SGSN and GGSN are the two end-points of the tunnel. At the MS side, a PDP context is identified by a network service access point identifier (NSAPI). The MS uses the appropriate NSAPI for subsequent data transfers to identify a PDN.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

121

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

  

On the other hand, the SGSN and GGSN use the TID to identify transfers with respect to a specific MS. Figure 7.10 illustrates an example of an MS with two PDP contexts activated. The MS uses NSAPI-1 for its data transfers with PDN1. The corresponding tunnel is identified by TID-1. Similarly, NSAPI-2 is used by the MS to connect to PDN2. The tunnel identifier in this case is TID-2.

Figure 7.10: A MS with PDP contexts active.   

Depending on the GPRS implementation, an MS can be assigned static or dynamic addresses. For instance, the operator can assign a permanent (static) PDP address to the MS or choose to assign a different address for each PDP context activated dynamically. Also, a visited network may assign dynamically an address to the MS for each PDP context activated.

Packet switching in GPRS Once the MS has attached, data activities can proceed with a PDN provided that a PDP session has been established with the PDN in question.  GPRS data are encapsulated and tunneled between the MS and the PDNs transparently through the GPRS network.  Sub-network dependent convergence protocol (SNDCP) provides compression and segmentation mechanisms and ensures the transfer of data packets between the MS and the SGSN.  GTP (GPRS tunneling protocol) manages tunneling of user packets between the SGSN and the GGSN. Data can also be transferred in a protected mode and monitored by retransmission protocols.  For a given active PDP context, data transfer can be mobile-originated (MO) or mobile-terminated (MT). The activation of the PDP context can be however MS-initiated or network requested and the transfer mode is independent of this fact.  MO packets are forwarded by the SGSN to the appropriate GGSN through the established GTP tunnel.  The GGSN then delivers the packets to the PDN. For MT transmissions, when the GGSN receives data packets for the MS, it identifies the SGSN that is currently serving it, and tunnels the packets.  The SGSN then delivers the packets to the MS in the appropriate cell. When the MS is in Standby, if it is paged in the appropriate RA and it responds to the page, it moves to the Ready state in order to receive the packets. Figure 7.11 illustrates an example of data routing for two active PDP contexts.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

122

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 7.11: End-to-end packet switching for a MS in the home network. Data routing for a mobile MS As a result of mobility, GPRS requires a mechanism to continue forwarding the packets to the MS when it enters a new SGSN service area. In GPRS, a buffering mechanism controlled by a timer in the old SGSN is used. This is how it works:  as soon as the old SGSN responds to the new SGSN with the MS context information, it starts a timer and starts tunneling buffered PDUs received from the GGSN to the new SGSN.  The old SGSN stops forwarding packets to the new SGSN when the timer expires.  The new SGSN buffers the packets received until the inter-SGSN routeing update procedure is completed and then delivers them. When the MS is roaming in a visited PLMN, it can still access GPRS services, perform mobility management functions, activate PDP contexts, and send and receive packets under the following conditions: o roaming agreements exist between the home PLMN and the visited PLMN, o the subscription allows for the requested services in the visited PLMN, o the presence of a border gateway between the home and the visited PLMNs to ensure the transport of signaling and data between both PLMNs. 7.7 GSM/GPRS SERVICE INTERACTIONS GPRS has a special network mode of operation (Mode I) allowing mobility management coordination between GSM and GPRS through the Gs interface. Interactions between GSM and GPRS services imply the need for interactions between the SGSN and the MSC/VLR and for the optional Gs interface between the two to be supported. An association is maintained between the SGSN and the MSC/VLR that are currently serving the MS so that paging for circuit-switched calls can be achieved via the SGSN when the MS is camped on the packet side. The SGSN is identified by a number that is communicated to the MSC/VLR during a mobility management update procedure. Similarly, the VLR is identified by a number that is derived from the RA identifier by the SGSN using a mapping table. The SGSN has to keep track of the VLR number while the latter stores the MS class and the SGSN number. Among the benefits of such association is the economy of radio resources. For instance, when an MS is already attached to the GPRS network (GPRS attached), it can request to attach to the GSM network (IMSI-attach) via the SGSN. A combined GPRS/IMSI attach is also possible and will be issued via the SGSN which initiates the association with the MSC/VLR. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

123

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

GSM location update procedures and paging for circuit-switched calls can be accomplished via the SGSN. “The association is updated every time the MS changes SGSN or MSC/VLR service areas and is removed at GPRS or IMSI detach”. Co-ordination of location area and routeing area updates   

The MS monitors the RA and LA every time it enters a new cell. If the MS is both GPRS and IMSI attached and the LA is changed, then it may use the RA update request to include the LA change. The SGSN forwards the LA update to the MSC/VLR.

In the example shown in Figure 7.12 and explained in Table 7.4, an MS moves from LA2 and crosses LA4 to finally reach LA5. 

 

During this transition, the MS has actually covered RA3, RA4, RA8, and RA11. Assuming that the MS is of class A or B and is in GSM Idle state, several possible location management procedures are required and are described in Table 8.4. Note that in GSM, no LA updates are performed during the dedicated state. Another possible scenario not shown in Table 7.4 is the case of combined Intra-SGSN RA/LA update. This occurs when the MS switches MSC/VLR but remains within the service area of the same SGSN. In this case a new association between the new MSC/VLR and the SGSN is created and the old association with the old MSC/VLR is removed. Note that the illustration of Figure 7.12 does not consider such scenario for simplicity but some networks may support it.

Figure 7.12: Location area and routeing area updates.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

124

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 7.4: Examples of location management procedures

Circuit-switched paging via the SGSN By maintaining the association between MSC/VLR and SGSN, the MSC can page the MS for incoming circuit-switched calls via the SGSN (Figure 7.13).

125

Figure 7.13: Circuit-switched page via the SGSN. Paging via the SGSN is more efficient because the SGSN will page in either a cell or a routeing area which is always contained in a GSM location area. The MS needs to listen therefore to only one paging channel, and the paging area is optimized. Note that the MS can be already engaged in a GPRS data transaction which may require extra handling depending on the class of the MS. The special case of class-B MS A class-A MS can engage in GPRS and GSM calls simultaneously which means that it can continue receiving packets while carrying a GSM voice call. Class-B on the other hand is a special case since it can be both IMSI/GPRS attached but can only service one of them at a time. In the case where the MS is already engaged in a GPRS data transaction, an incoming GSM call requires the GPRS packet activity to be suspended until the voice call is terminated. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Point-to-point short message service (SMS) GPRS is also defined to support the GSM SMS service. The idea is to allow GPRS attached mobiles to send and receive short messages via the SGSN and over the GPRS radio channels.  This feature aims to further optimize the use of radio resources and to give operators more flexibility in terms of SMS delivery.  An additional interface (Gd interface) has been defined for the purpose of SMS support and connects the SGSN to SMS gateway nodes. 7.8 GPRS EFFICIENCY Figure 7.14 shows an idealized comparison of GPRS and circuit-switched data services for typical Internet browsing. In this context, throughput is the average throughput that a user experiences as he or she downloads information from the Internet. In the case of GPRS, fewer active users implies that each user has access to more bandwidth. As the number of active users grows, the bandwidth allocated to each user decreases. Compare this to circuit switched service, where fixed bandwidth is allocated to a limited number of users. Compared with circuit-switched connections, GPRS offers superior performance to applications like Internet browsing.  Due to bursty user behavior (users suddenly require lots of bandwidth, then nothing, then lots of bandwidth, and so forth), GPRS can serve more users than ordinary circuit-switched services.  On the other hand, GPRS offers nonbursty applications the same level of service- in terms of throughput- as circuit-switched data. Obviously, in evaluating efficiency, the user traffic model plays a central role.

Figure 7.14: Idealized comparison of GPRS and circuit-switched data services. 7.9 LIMITATIONS OF GPRS Despite its potential, GPRS remains constrained by the following limitations. 

GPRS transmission rates are much lower than in theory. To achieve the theoretical maximum of about 170 kbit/s would require allocating eight time slots to a single user which is not likely to be allowed by network operators. Even if this maximum allocation was allowed, the GPRS terminals may be constrained by the number of time slots they can handle.



GPRS relies on packet switching which means that data packets can traverse different routes and then be reassembled in their final destination leading to potential transit delays affecting the QoS.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

126

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



GPRS relies also on re-transmission and data integrity protocols to ensure that data packets transmitted over the radio air interface are not lost or corrupted. This can affect even further the transit delay problem.



The protocols between the BSS and the SGSN support mainly asynchronous data transfer applications making it a challenge to implement real-time interactive traffic.

127

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 08 ENHANCED DATA RATES FOR GSM EVOLUTION 8.1 OVERVIEW EDGE is the next step in the evolution of GSM and IS-136. The objective of the new technology is to increase data transmission rates and spectrum efficiency and to facilitate new applications and increased capacity for mobile use. With the introduction of EDGE in GSM phase 2+, existing services such as GPRS and high-speed circuit switched data (HSCSD) are enhanced by offering a new physical layer. EDGE is introduced within existing specifications and descriptions rather than by creating new ones. GPRS allows data rates of 115 kbps and, theoretically, of up to 160 kbps on the physical layer. EGPRS is capable of offering data rates of 384 kbps and, theoretically, of up to 473.6 kbps. A new modulation technique and error-tolerant transmission methods, combined with improved link adaptation mechanisms, make these EGPRS rates possible. This is the key to increased spectrum efficiency and enhanced applications, such as wireless Internet access, e-mail and file transfers. Hence, Three primary attributes are addressed by EDGE are as follows:   

High order modulation scheme (8-PSK) 3 bits per symbol instead of 1 Radio link adaptation Agility at radio layer to exploit local radio conditions Incremental redundancy Reduces re-transmissions of radio blocks by hybrid coding scheme

EDGE can be introduced in two ways:  as a packet-switched enhancement for general packet radio service (GPRS), known as enhanced GPRS or EGPRS, and  as a circuit-switched data enhancement called enhanced circuit-switched data (ECSD). However, we shall mainly discuss the packet-switched enhancement, EGPRS. 8.2 GPRS AND EGPRS ARCHITECTURE Regarded as a subsystem within the GSM standard, GPRS has introduced packet-switched data into GSM networks. Many new protocols and new nodes have been introduced to make this possible. EDGE is a method to increase the data rates on the radio link for GSM. Basically, EDGE only introduces a new modulation technique and new channel coding that can be used to transmit both packetswitched and circuit-switched voice and data services. EDGE is therefore an add-on to GPRS and cannot work alone. GPRS has a greater impact on the GSM system than EDGE has. “By adding the new modulation and coding to GPRS and by making adjustments to the radio link protocols, EGPRS offers significantly higher throughput and capacity”. “GPRS and EGPRS have different protocols and different behavior on the base station system side. However, on the core network side, GPRS and EGPRS share the same packet-handling protocols and, therefore, behave in the same way”.   

Reuse of the existing GPRS core infrastructure (serving GRPS support node/gateway GPRS support node) emphasizes the fact that EGPRS is only an “add-on” to the base station system and is therefore much easier to introduce than GPRS (Figure 8.1). In addition to enhancing the throughput for each data user, EDGE also increases capacity. With EDGE, the same time slot can support more users.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

128

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

  

This decreases the number of radio resources required to support the same traffic, thus freeing up capacity for more data or voice services. EDGE makes it easier for circuit-switched and packet-switched traffic to coexist while making more efficient use of the same radio resources. Thus in tightly planned networks with limited spectrum, EDGE may also be seen as a capacity booster for the data traffic.

129

Figure 8.1: EGPRS introduces changes to GPRS only on the base station system part of the network. 8.3 EDGE TECHNOLOGY Although GPRS and EDGE share the same symbol rate, the modulation bit rate differs. EDGE can transmit three times as many bits as GPRS during the same period of time. This is the main reason for the higher EDGE bit rates. Figure 8.2 compares the basic technical data of GPRS and EDGE.

Figure 8.2: GPRS and EDGE: A comparison of technical data. (Note that 8PSK: 8-phase shift keying; GMSK: Gaussian minimum shift keying; MCS: Modulation coding scheme). EDGE Modulation Technique Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The modulation type that is used in GSM is the Gaussian minimum shift keying (GMSK), which is a kind of phase modulation. This can be visualized in an I/Q diagram that shows the real (I) and imaginary (Q) components of the transmitted signal (Figure 8.3). Transmitting a zero bit or one bit is then represented by changing the phase. Every symbol that is transmitted represents one bit; that is, each shift in the phase represents one bit. To achieve higher bit rates per time slot than those available in GSM/GPRS, the modulation method requires change. “EDGE is specified to reuse the channel structure, channel width, channel coding and the existing mechanisms and functionality of GPRS and HSCSD”.   

The modulation standard selected for EDGE, 8-phase shift keying (8PSK), fulfills all of those requirements. 8PSK modulation has the same qualities in terms of generating interference on adjacent channels as GMSK. This makes it possible to integrate EDGE channels into an existing frequency plan and to assign new EDGE channels in the same way as standard GSM channels.

The 8PSK modulation method is a linear method in which three consecutive bits are mapped onto one symbol in the I/Q plane.  The symbol rate, or the number of symbols sent within a certain period of time, remains the same as for GMSK, but each symbol now represents three bits instead of one. The total data rate is therefore increased by a factor of three.  Note that the distance between the different symbols is shorter using 8PSK modulation than GMSK. Shorter distances increase the risk for misinterpretation of the symbols because it is more difficult for the radio receiver to detect which symbol it has received. Under good radio conditions, this does not matter.  Under poor radio conditions, however, it does. The “extra” bits will be used to add more error correcting coding, and the correct information can be recovered. Only under very poor radio environments is GMSK more efficient. Therefore, the EDGE coding schemes are a mixture of both GMSK and 8PSK.

Figure 8.3: I/Q diagram showing EDGE modulation benefits. Coding Schemes

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

130

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

For GPRS, four different coding schemes, designated CS1 through CS4, are defined. Each has different amounts of error-correcting coding that is optimized for different radio environments. For EGPRS, nine modulation coding schemes, designated MCS1 through MCS9, are introduced. These fulfill the same task as the GPRS coding schemes. The lower four EGPRS coding schemes (MSC1 to MSC4) use GMSK, whereas the upper five (MSC5 to MSC9) use 8PSK modulation. Figure 8.4 shows both GPRS and EGPRS coding schemes, along with their maximum throughputs. “GPRS user throughput reaches saturation at a maximum of 20 kbps with CS4, whereas the EGPRS bit rate continues to increase as the radio quality increases, until throughput reaches saturation at 59.2 kbps”. Note that both GPRS CS1 to CS4 and EGPRS MCS1 to MCS4 use GMSK modulation with slightly different throughput performances.  This is due to differences in the header size (and payload size) of the EGPRS packets. This makes it possible to resegment EGPRS packets.  A packet sent with a higher coding scheme (less error correction) that is not properly received, can be retransmitted with a lower coding scheme (more error correction) if the new radio environment requires it.  This resegmenting (retransmitting with another coding scheme) requires changes in the payload sizes of the radio blocks, which is why EGPRS and GPRS do not have the same performance for the GMSK modulated coding schemes.  Resegmentation is not possible with GPRS. 131

Figure 8.4: Coding schemes for GPRS and EGPRS (user data rate). (Note that 8PSK: 8-phase shift keying; CS: Coding scheme; EGPRS: Enhanced GPRS; GMSK: Gaussian minimum shift keying; MCS: Modulation coding scheme). EGPRS Link Controlling Function To achieve the highest possible throughput over the radio link, EGPRS uses a combination of two functionalities: link adaptation and incremental redundancy. Compared to a pure link adaptation solution, this combination of mechanisms significantly improves performance. Link Adaptation

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Link adaptation uses the radio link quality, measured either by the mobile station in a downlink transfer or by the base station in an uplink transfer, to select the most appropriate modulation coding scheme for transmission of the next sequence of packets. For an uplink packet transfer, the network informs the mobile station which coding scheme to use for transmission of the next sequence of packets.  The modulation coding scheme can be changed for each radio block (four bursts).  But a change is usually initiated by new quality estimates. The practical adaptation rate is therefore decided by the measurement interval. There are three families: A, B, and C. Within each family, there is a relationship between the payload sizes, which makes resegmentation for retransmissions possible.

132

Figure 8.5: Modulation and coding schemes. (Note that 8PSK: 8-phase shift keying; GMSK: Gaussian minimum shift keying; MCS: modulation coding scheme) Incremental redundancy Incremental redundancy initially uses a coding scheme, such as MCS9, with very little error protection and without consideration for the actual radio link quality.  When information is received incorrectly, additional coding is transmitted and then soft combined in the receiver with the previously received information.  Soft-combining increases the probability of decoding the information. This procedure will be repeated until the information is successfully decoded.  This means that information about the radio link is not necessary to support incremental redundancy.  For the mobile stations, incremental redundancy support is mandatory in the standard.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Figure 8.6: Incremental redundancy. (Note that 8PSK:8-phase shift keying; GMSK: Gaussian minimum shift keying; MCS: Modulation coding scheme). Packet Handling Another improvement that has been made to the EGPRS standard is the ability to retransmit a packet that has not been decoded properly with a more robust coding scheme. For GPRS, resegmentation is not possible. Once packets have been sent, they must be retransmitted using the original coding scheme even if the radio environment has changed. This has a significant impact on the throughput, as the algorithm decides the level of confidence with which the link adaptation (LA) must work. Below is an example of packet transfer and retransmission for GPRS (Figure 8.7). A. The GPRS terminal receives data from the network on the downlink. Due to a GPRS measurement report that was previously received, the link adaptation algorithm in the base station controller decides to send the next radio blocks (e.g., numbers 1 to 4) with CS3. During the transmission of these packages, the carrier-to-interference ratio (C/I) decreases dramatically, changing the radio environment. After the packets have been transmitted, the network polls for a new measurement report, including the acknowledged/unacknowledged bitmap that tells the network which radio blocks were received correctly. B. The GPRS handset replies with a packet downlink acknowledged/unacknowledged message containing the information about the link quality and the bitmap. In this scenario, it is assumed that packets 2 and 3 were sent erroneously. C. Based on the new link quality information, the GPRS link adaptation algorithm will adapt the coding scheme to the new radio environment using CS1 for the new packets 5 and 6. However, because GPRS cannot resegment the old packets, packets 2 and 3 must be retransmitted using CS3, although there is a significant risk that these packets still may not be decoded correctly. As a result, the link adaptation for GPRS requires careful selection of the coding scheme in order to avoid retransmissions as much as possible. With EGPRS, resegmentation is possible. Packets sent with little error protection can be retransmitted with more error protection, if required by the new radio environment. The rapidly changing radio environment has a much smaller effect on the problem of choosing the wrong coding scheme for the next sequence of radio blocks because resegmentation is possible. Therefore, the EGPRS link-controlling algorithm can be very aggressive when selecting the modulation coding schemes.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

133

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 8.7: Packet transfer and retransmission for GPRS. (Note that: ACK/NACK, acknowledged/unacknowledged; CS, Coding Scheme). 134

Interleaving To increase the performance of the higher coding schemes in EGPRS (MCS7 to MCS9) even at low C/I, the interleaving procedure has been changed within the EGPRS standard.  When frequency hopping is used, the radio environment is changing on a per-burst level.  Because a radio block is interleaved and transmitted over four bursts for GPRS, each burst may experience a completely different interference environment.  If just one of the four bursts is not properly received, the entire radio block will not be properly decoded and will have to be retransmitted.  In the case of CS4 for GPRS, hardly any error protection is used at all. With EGPRS, the standard handles the higher coding scheme differently than GPRS to combat this problem.  MCS7, MCS8 and MCS9 actually transmit two radio blocks over the four bursts, and the interleaving occurs over two bursts instead of four.  This reduces the number of bursts that must be retransmitted should errors occur.  The likelihood of receiving two consecutive error free bursts is higher than receiving four consecutive error free bursts.  This means that the higher coding schemes for EDGE have a better robustness with regard to frequency hopping.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 8.8: Interleaving. (Note that: CS: coding scheme; EGPRS: enhanced GPRS; MCS: modulation coding scheme). Impact of EGPRS on existing GSM/GPRS networks Due to the minor differences between GPRS and EGPRS, “The impact of EGPRS on the existing GSM/GPRS network is limited to the base station system. The base station is affected by the new transceiver unit capable of handling EDGE modulation as well as new software that enables the new protocol for packets over the radio interface in both the base station and base station controller. However, the core network does not require any adaptations”. Due to this simple upgrade, a network capable of EDGE can be deployed with limited investments and within a short time frame. 8.4 EGPRS ARCHITECTURE AND PROTOCOLS EGPRS does not bring about any direct architecture impacts.  The packet control unit may still be placed either in the base station, the base station controller or the GPRS support node, and  the central control unit is always placed in the base station. However, since the radio link control automatic repeat request function on the network side is located in the packet control unit, any delay introduced between the PCU and the radio interface will directly affect the radio link control acknowledged/unacknowledged round-trip times.  This, in turn, results in a higher risk of stalling the radio link control protocol.  To mitigate this risk and to allow the operator to optimize network behavior, the maximum radio link control automatic repeat request window size has been extended for EGPRS. The transmission plane protocol structure for GPRS is shown in Figure 8.9.  The protocols (user plane) that are influenced by the introduction of EDGE are shaded.  The protocols closest to the physical layer (the radio link control and mobile allocation channel) are most affected by EDGE.  There also are some minor modifications to the base station system GPRS protocol.  Apart from these changes, the rest of the protocol stack remains intact after the introduction of EDGE. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

135

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The introduction of EGPRS also has an impact on the control plane layers:  mobility management and  radio resource management. However, there is no impact on session management. The mobility management modifications are related to introducing information on EGPRS capabilities in the mobile station radio access capabilities information element. These capabilities include  the EGPRS multislot class,  the EDGE modulation capability and  the 8PSK power class. On the radio resource management layer, support for setting up and maintaining EGPRS temporary block flows is introduced as opposed to standard GPRS temporary block flows. Signaling supporting the  radio link control,  link quality control and  measurement procedures is also introduced.

136

Figure 8.9: Transmission plane protocol architecture. (Note that BSS, Base station system; BSSGP, BSS GPRS protocol; GGSN, Gateway GPRS support node; GTP, General telemetry processor; IP/X.25, Internet Protocol X.25; LLC, Low-layer capability; L1 and L2, memory caches; MAC, Mobile allocation control; MS, Mobile station; RF, Radio frequency; RLC, Radio link control; SGSN, Serving GPRS support node; SNDCP, Subnetwork-dependent convergence protocol; TCP, Transmission control protocol; UDP, User diagram protocol). 8.5 EGPRS BENEFITS Short-term benefits: Capacity and performance    

EGPRS introduces a new modulation technique, along with improvements to the radio protocol, that allows operators to use existing frequency spectrums (800, 900, 1800 and 1900 MHz) more effectively. The simple improvements of the existing GSM/GPRS protocols make EDGE a cost-effective, easyto implement add-on. Software upgrades in the base station system enable use of the new protocol; new transceiver units in the base station enable use of the new modulation technique. EDGE triples the capacity of GPRS.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

    

This capacity boost improves the performance of existing applications and enables new services such as multimedia services. It also enables each transceiver to carry more voice and/or data traffic. EDGE enables new applications at higher data rates. This will attract new subscribers and increase an operator’s customer base. Providing the best and most attractive services will also increase customer loyalty.

Mid-term benefits: Complementary technology “EDGE and WCDMA are complementary technologies that together will sustain an operator’s need for third generation network coverage and capacity nationwide”.  Enhancing a GPRS network is accomplished through evolution with EDGE within the existing spectrum and by deploying WCDMA in the new frequency band.  Rolling out the two technologies in parallel enables faster time to market for new high-speed data services as well as lower capital expenditures.  EDGE is designed to integrate into the existing network.  The installed base evolves; it is not replaced or built from scratch, making implementation seamless.  Fast, easy rollout means shorter time to market, which in turn can lead to increased market share.  With EDGE, operators can offer more wireless data applications, including o wireless multimedia, o e-mail, web infotainment and o positioning services, for both consumer and business users.  Subscribers will be able to browse the Internet on their mobile phones, personal digital assistants or laptops at the same speed as on stationary personal computers. Long-term benefit: Harmonization with WCDMA EDGE can be seen as a foundation toward one seamless GSM and WCDMA network with a combined core network and different access methods that are transparent to the end user.

Figure 8.10: One seamless GSM and WCDMA network.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

137

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 09 EDGE EVOLUTION ENHANCEMENTS 9.1 INTRODUCTION EDGE and EDGE Evolution are designed with both backwards compatibility and efficiency in mind. Both steps of the technology substantially improve performance and efficiency in the GSM network while protecting the architecture and keeping as much existing equipment as possible in use. These benefits make upgrades more cost efficient for operators with existing GSM infrastructure. EDGE Evolution Performance Boost To improve service performance in general, and facilitate conversational multimedia services, a number of enhancements to EDGE have been standardized by the 3GPP. Known collectively as EDGE Evolution, these were included in Release 7 of the 3GPP standard. Peak bit rates of up to 1Mbps and typical bit rates of 400kbps can be expected. Round-trip times will be less than 80ms and spectrum efficiency will be more than twice as good as today. EDGE Evolution can be gradually introduced as software upgrades, taking advantage of the installed base. With EDGE Evolution, end users will be able to experience mobile internet connections corresponding to a 500kbps ADSL service. EDGE Evolution will improve service performance and enable more efficient radio bearers. Different services may have varying performance requirements in different areas, but EDGE Evolution is expected to improve the perceived performance across all services by:    

Reducing latency to improve the user experience of interactive services and also to enhance support for conversational services such as multimedia telephony. Increasing peak and mean bit rates, to improve best-effort services such as web browsing or music downloads. Improving spectrum efficiency, which will particularly benefit operators in urban areas where existing frequency spectrum is used to its maximum extent – traffic volume can be increased without compromising service performance or degrading perceived user quality. Boosting service coverage, for example by reducing interference or allowing more robust services. Increased terminal sensitivity improves coverage in the noise limited scenario.

Implementing EDGE Evolution The installed base of GSM/EDGE equipment is very large, so great care has been taken to ensure that the impact of EDGE Evolution on base station hardware is minimized. The different enhancements may be gradually – and to some extent independently – introduced in the network, most of them as software upgrades. Current network architecture remains unchanged. Handsets will require more extensive modifications, but are replaced at a much higher rate. A large number of handset vendors are foreseen to adhere to EDGE Evolution, and handsets with increasing levels of EDGE Evolution functionality are expected to be available from 2010. Figures 9.1 and 9.2 show examples of the increased peak bit rates and spectrum efficiency provided by GPRS, EDGE and different stages of EDGE Evolution implementation. Performance in a live network has been measured in order to ensure that EDGE Evolution provides benefits in real situations. Quality measurements have been collected in a number of typical cells in operators’ networks and the results show that EDGE Evolution will significantly improve bit rates in the whole cell.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

138

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure9.1: Peak bit rates in downlink for GPRS, EDGE and different stages of EDGE Evolution Figure 9.3 shows the cumulative distribution of the time slot bit rate in an urban cell with medium quality. EDGE Evolution increases the average bit rate by 86 percent (from 42 to 78kbps) compared with EDGE.

139

Figure 9.2: Relative spectrum efficiency for GPRS, EDGE and different stages of EDGE Evolution.

Figure 9.3: Bit rate distribution per time slot in a rural cell for EDGE and EDGE Evolution.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Latency Reduction EDGE Evolution makes substantial improvements in latency and perceived delay through reduced Transmission Time Interval (TTI) and additional protocol enhancements. Radio blocks are currently transmitted over four consecutive bursts on one time slot using a TTI of 20ms. Reducing the TTI to 10ms improves latency substantially, to below 80ms. The four bursts are then transmitted on more than one time slot (parallel time slots on two carriers or dual time slots on one carrier).

Figure 9.4: Lower latency with reduced TTI. Reduced TTI Latency has a major influence on user experience. In particular, conversational services, such as VoIP and video telephony, require low latency. Other services that benefit from low delay are gaming and applications with extensive handshaking, such as e-mail. It is difficult to substantially improve latency without reducing the transmission time interval (TTI). The roundtrip time (RTT) in advanced GSM/EDGE networks with an Ericsson base station subsystem (BSS) is 150ms. This figure includes network delays but not retransmissions over the radio interface. Radio blocks are transmitted over four consecutive bursts on one timeslot using a 20ms TTI. Figure 9.5 illustrates the data flow with a 20ms TTI. Reducing the TTI improves latency substantially and immediately. To reduce TTI, one can either use fewer than four bursts (smaller radio blocks) or transmit all four bursts on more than one timeslot (for example, parallel timeslots on two carriers). Note that, Ericsson estimates that reducing the TTI from 20ms to 10ms will reduce the roundtrip time from 150ms to 100ms. Faster Feedback To help the transmitter better understand the radio environment, feedback information is sent via the radio link control (RLC) protocol over the air. The RLC protocol typically runs in acknowledged mode, which requires the retransmission of lost radio blocks. Although feedback information is crucial for efficient transmission over the radio interface, it is also time-consuming. The procedure requires the receiver to periodically send (on request)  acknowledgements of radio transmissions; and  information about the current radio environment. Faster feedback enables the transmitter to retransmit lost data earlier and makes radio transmission more efficient. By putting more stringent requirements on reaction times and by introducing support for immediate response to unsuccessful radio transmissions, one can ensure that lost radio blocks are retransmitted much earlier, which reduces latency. One can reduce latency even further by combining faster feedback with reduced TTI (Figure 9.6).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

140

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 9.5: Data flow diagram with TTI = 20ms.

141

Figure 9.6: Example of transfer using reduced TTI and faster feedback enhancements. 9.2 INCREASED BIT RATES AND IMPROVED EFFICIENCY Dual Carriers The most obvious improvement to peak bit rates is through the introduction of dual carriers in the downlink, increasing the carrier bandwidth available above 200kHz. This overcomes the inherent limitation of the narrow channel bandwidth of GSM.  The introduction of dual carriers doubles the available bandwidth (to 400 kHz) as well as the practical peak bit rate.  Using dual carriers and five timeslots on each carrier provides bitrates of almost 600kbps, with no other changes to EDGE.  Using two radio-frequency carriers requires two receiver chains in the downlink, as shown in Figure 9.7.  Using two carriers enables the reception of more than twice as many radio blocks simultaneously. Having a second receiver chain also permits the mobile device to use one receive chain for neighbour cell monitoring, which then permits the mobile device to receive up to five timeslots in the downlink instead of four, as shown in Figure 9.8. Alternatively, the original number of radio blocks can be divided between the two carriers. This eliminates the need for the network to have contiguous timeslots on one frequency.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 9.7: Evolved EDGE Two-Carrier Operation.

142

Figure 9.8: Evolved EDGE Neighbor Cell Monitoring. Channel capacity with dual-carrier reception improves greatly, not by increasing basic efficiencies of the airinterface but because of statistical improvement in the ability to assign radio resources, which increases trunking efficiency.  As network loading increases, it is statistically unlikely that contiguous timeslots will be available.  With today’s EDGE devices, it is not possible to change radio frequencies when going from one timeslot to the next.  However, with an Evolved EDGE dual receiver this becomes possible, thus enabling contiguous timeslots across different radio channels.  Figure 9.9 shows a dual-radio receiver approach optimizing the use of available timeslots. (“Rx1” refers to receiver 1, “Rx2” refers to receiver 2, “NCM” refers to neighbour cell monitoring, and “M2” refers to receiver 2 doing system monitoring.).  Through intelligent selection, dual-carrier receiver architecture can support either dual-carrier reception or mobile-station receive diversity, depending on the operating environment.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 9.9: Optimization of Timeslot Usage Example. Higher-Order Modulation, Turbo Codes and Increased Symbol Rate The addition of higher order modulation schemes enhances EDGE network capacity with little capital investment by extending the range of the existing wireless technology.  More bits per symbol mean more data transmitted per unit time. This yields a fundamental technological improvement in information capacity and faster data rates.  Use of higher order modulation exploits localized optimal coverage circumstances, thereby taking advantage of the geographical locations associated with probabilities of high C/I ratio and enabling very high data transfer rates whenever possible. Higher average and peak bit-rates and improved spectrum efficiency are achieved through more advanced modulation, more efficient channel coding and an increased symbol rate (in practice, increasing the carrier bandwidth). Two different levels of support for higher order modulation are defined for both the uplink and the downlink. In the uplink, the first support level includes GMSK, 8-PSK, and 16 QAM at the legacy symbol rate. This level of support reuses MCSs 1 through 6 from EGPRS and adds five new 16 QAM modulated MCSs. The second support level in the uplink includes QPSK, 16 QAM, and 32 QAM modulation as well as a higher (1.2x) symbol rate. MCSs 1 through 4 from EGPRS are reused, and eight new MCSs are added. The first downlink support level adds 8-PSK, 16 QAM, and 32 QAM at the legacy symbol rate. MCSs 1 through 4 are reused, and eight new MCSs are added. The second downlink support level includes QPSK, 16 QAM, and 32 QAM modulations at a higher (1.2x) symbol rate. MCSs 1 through 4 are reused, and eight new MCSs are defined. The combination of Release 7 EDGE Evolution enhancements shows a dramatic potential increase in throughput. For example, in the downlink, a Type 2 mobile device (one that can support simultaneous transmission and reception) using HTCS-8-B as the MCS and a dual-carrier receiver can achieve the following performance: Highest data rate per timeslot (layer 2) = 118.4 kbps Timeslots per carrier = 8 Carriers used in the downlink = 2 Total downlink data rate = 118.4 kbps X 8 X 2 = 1894.4 kbps75 This translates to a peak network rate close to 2 Mbps and a user-achievable data rate of well over 1 Mbps.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

143

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 9.10: Top left: Gaussian minimum-shift keying (GMSK). Top right: Octonary phase shift keying (8PSK). Bottom left: 16QAM (16-level constellation). Bottom right: 32QAM (32-level constellation). Table 9.1: Uplink Modulation and Coding Schemes.

144

Table 9.2: Uplink Modulation and Coding Schemes with Higher Symbol Rate.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Table 9.3: Downlink Modulation and Coding Schemes. 145

Table 9.4: Downlink Modulation and Coding Schemes with Higher Symbol Rate.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Dual-Antenna Terminals

146

A challenge associated with radio communication is that the strength of received radio signals varies rapidly relative to the position of the transmitter, the receiver, and other objects that scatter signals. This phenomenon, known as fading, can make a signal too weak to be captured. One other challenge is that other transmitters (terminals or base stations) in the vicinity can cause interference. Various techniques, such as channel coding, frequency hopping, and selective retransmission schemes, are used to combat fading and interference. Dual-receive-antenna systems are an efficient weapon against fading. To date, however, these have only been used on base station receivers. Figure 9.11 shows how two antennas mounted on a terminal, separated in space, polarization, or both, can receive two signals of the same transmission with different fading characteristics. This increases the probability that at least one of the signals will be strong enough (with some margin above the receiver noise floor) to be captured.

Figure 9.11: Received signal strength on two antennas in a mobile terminal. Moreover, by combining two signals, one can sometimes capture a transmission that would otherwise have been altogether too weak. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Dual-antenna solutions can also be used to efficiently handle interference. Obviously, the desired signal as well as all other received interference is subject to fading on the two antennas. But by combining the signals, one can cancel out interference by taking into account the instantaneous attenuation of the different signals due to fading (denoted a, b, c and d, Figure 9.12). This technique is known as interference cancellation. Experiments and computer simulations show that dual-antenna solutions in GSM terminals yield substantial improvement. In situations with limited coverage (that is, when the signal is too weak to be received correctly), dual-antenna terminals can cope with signal levels 6dB below (or one-fourth) that of single-antenna terminals. Moreover, dual-antenna terminals can handle almost 10dB (or ten times) more interference.

147

Figure 9.12: Dual receive antennas can be used for interference cancellation. The received signal on each antenna is the sum of the desired signal and interference, weighted with different attenuation factors due to fading (denoted a, b, c and d). Mobility Enhancements The functionality needed to handle user mobility is a fundamental component in any cellular network. Services must be maintained as users move from one cell to another, and a good or adequate radio environment must exist for the establishment of new sessions. Seen in terms of spectrum efficiency, each active user should always be served by the most appropriate (closest) base station. In advanced GSM/EDGE networks that employ tight frequency reuse, being connected to the most appropriate base station is essential for achieving good spectrum efficiency. In tight frequency-reuse scenarios, user mobility can compromise service quality, particularly in cases that involve an inhomogeneous cell plan, fast-moving users, or tight reuse of the broadcast control channel (BCCH). User mobility is managed in three general steps (in both active and idle modes for voice and data but with different processes and requirements in each case): measurements are taken to determine whether or not cell change is necessary; processing of the measurements (to reach this decision); and the actual cell change. In idle mode, the main limitation of managing user mobility is that the measurements of neighboring and serving cells are not stringent enough for environments with tight frequency reuse. Consequently, sessions are sometimes initiated in a suboptimum cell, creating extra interference and leading to service interruptions (dropped calls). Putting more stringent requirements on measurements of the received signal strength would eliminate this limitation. In active mode, faster and more accurate measurements of neighboring cells would improve performance. At present, mobile terminals identify neighboring cells separately and infrequently by decoding the base station identity code (BSIC). Furthermore, they solely measure the total strength of the received signal. This means that if several nearby base stations use the same frequency, the contribution from each base station cannot be resolved. Ericsson proposes to introduce simultaneous identification of neighboring cells (BSIC information) and measurements of signal strength on all bursts over the BCCH carrier. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

This would significantly reduce the risk of incorrect handover, resulting in substantially fewer dropped calls. It would also speed up the process of detecting new, strong, neighboring cells. Mobility enhancements apply equally well to voice and data services (although the detailed effects differ). They also improve service continuity, for example, by means of faster and more accurate handover between GSM/EDGE and WCDMA networks (Figure 9.13) and enhanced handover functionality in dualtransfer mode (DTM).

Figure 9.13: Mobility enhancements. 148

9.3 SERVICE COVERAGE Dual antenna terminals can also improve service coverage. With two antennas and efficient combination methods, weaker signal transmissions can be captured. Around 3Db less (roughly 50 percent) signal power is needed to provide service, enabling larger cells or lower output power.

9.4 SUMMARY OF EDGE EVOLUTION A summary of the benefits of each individual enhancement follows below: 

 





Dual-antenna terminals improve capacity and mean bit rates substantially, especially when used to cancel interference in the downlink. Furthermore, receiver sensitivity is vastly improved, which improves coverage (as defined in the noise-limited case). Individual users benefit immediately. System gains are dependent on terminal penetration. Multicarrier EDGE yields increased bandwidth. This is manifested by increased peak and mean bit rates both over the uplink and downlink. Operators can retain the current channel structure, introducing complexity on a step-by-step basis as demand for bandwidth grows. Mobility enhancements increase mean bit rates and spectrum efficiency because users are always served by the most appropriate base station. Service continuity is also improved thanks to faster and more accurate handovers within GSM/EDGE networks and between GSM/EDGE and WCDMA networks. Reduced TTI and faster feedback reduce latency and improve overall service performance. Sessions that involve interactive data transfers will on average take less time. The enhancements also significantly reduce the mouth-to-ear delay and improve hardware efficiency for conversational services (for example, VoIP). Higher-order modulation used in combination with turbo coding substantially increases peak and mean bit rates. Service coverage is also improved thanks to increased robustness (when, for

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

example, MCS-8 and MCS-9 are realized with 16QAM instead of 8-PSK, which is used in the existing EDGE standards).

Figure 9.14 shows an example of different bit rates in a cell, as different features are introduced. It shows how higher-order modulation and dual carriers improve peak bit rates, while higher-order modulation, turbo codes and interference cancellation with dual antennas increase bit rates at the cell border.

149

Figure 9.14: Examples of bit rate improvements in different parts of a cell.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

CHAPTER 10 CODE DIVISION MULTIPLE ACCESS 10.1 MULTIPLE ACCESS TECHNIQUE Consider four communications entities: space, frequency, time and code. More than one communication signal can be carried by a single channel if any one of the above entity is different for those signals. That is, the receiver side can separate the desired signal from all other signals. 

Space Division Multiple Access where same time, same frequency, same coding for all signals but different space for each of them is employed.



Time Division Multiple Access where same space, same frequency, same coding for all signals but different time for each of them is employed.



Frequency Division Multiple Access where same space, same time, same coding for all signals but different frequency for each of them is employed.



Code Division Multiple Access where same time, same space, frequency for all signals but different code for each of them is employed. 150

Figure 10.1: Multiple access schemes. 10.2 CDMA PRINCIPLE In a code-division multiple-access system, which is also called spread-spectrum multiple access (SSMA) system, all the active users share the same frequency band and transmit at the same time.  Each user is distinguished from the others by a code assigned to him/her.  If the codes are orthogonal, then the signals from other users can be separated and filtered out at the receiver. Otherwise, the signals from other users become interference.  The codes assigned to the users are PN sequences.  It is not possible to generate a large set of orthogonal sequences. Therefore, interference from other users becomes a major impairment.  When the total interference reaches a level that the performance of the receiver is unacceptable, the number of active users cannot be further increased. Therefore, such multiple-access systems are interference limited. The basic concept of CDMA can be summarized as follows. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

   

Transmitted signal occupies a bandwidth much larger than the original signal, The bandwidth is spread by means of a PN code independent of the data, The receiver synchronizes to the code to recover the data, Many signals share the same times and frequencies but independent codes.

The CDMA technique codes each bit of data into a pattern looks like many tiny bits, called Chips (see the figure below). The chip pattern (chipping sequence) is the ‘Code’. Each transmitting station uses separate code. If the receiving station uses the same code then it can recover this signal from a mix of many signals (at same frequency, time and space).

151

Figure 10.2: CDMA technique. Since the coded signal looks like higher bit-rate signal (7 times higher in the above example) the bandwidth of the resultant signal is expected to be higher. In the above example, if the original signal bandwidth is 1 MHz then the coded signal’s bandwidth is 7 MHz. However, this seven bandwidth is not used just this signal. It is shared by many others (bandwidth per user is lower enough). The following figures show how CDMA signal is generated in the transmitter and recovered in the receiver. The CDMA technology is wining over the TDMA (just like how TDMA won over FDMA). Apart from its comparable bandwidth efficiency with TDMA it has the following two major advantages. Security CDMA is coded. If the code is secret the communication is secured.

Figure 10.3: CDMA signal generation in the transmitter and recovery in the receiver. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

No Hard Limit of Capacity TDMA defines a fixed number of time-slot (and that’s it - no more, no less) and same quality of service for all slots. In contrast, CDMA can accommodate up to 2N users, where N is code length (number of chips per code). If the code length is, for example, 16 then the number of user is about 64 thousands. But this is not that high since it is a requirement to keep the codes different in a signification number of bits (not just one or two). Otherwise, the receiver may not be able to detect the signal without too many errors. 

That is, higher code separation is related to better signal reception.



The response to the question of how much separation is good enough is like what voice quality is good enough for you in a poor coverage area.



The fact is there is no hard threshold. This is the key to the flexibility we are talking about. If a network operator reduces its quality it can allocate more codes and hence more users using the same bandwidth (Just imagine a poorer quality at peak hours and better quality at other hours). 10.3 SPREDING AND DESPREADING OPERATION

The bandwidth of the coded signal (transmitted signal after spreading the original data) is much larger than the original data signal. The bandwidth is spread by means of a spreading code which is random in appearance (pseudo-noise-like or PN sequence) and independent of the data. 

The spreading operation for this particular example is the multiplication of each user data bit with a sequence of 8 code bits, called chips. The resulting spread data is at a rate of 8 X R, where R is the original data rate. The number of code bits used in spreading operation is called the spreading factor (here it is 8).



This wideband signal is transmitted across a wireless channel to the receiving end. Because of the widening of original data signal by a factor equal to the Spreading factor, CDMA systems are called spread spectrum (SS) systems.



The receiver synchronizes to the code to recover the data, and the operation is called despreading. In despreading, the spread user data/chip sequence is multiplied with same code chips (as used during spreading at the transmitting end bit duration by bit duration basis. This results the recovery of the original data with rate R.



Note that more than one user generated signals may share the same time and frequency resources but use independent codes to one another.



Figure 10.4 illustrates the spreading and despreading operation. The example uses BPSK modulated user data and the user data bits assuming the values of either +1 or -1.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

152

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 10.4: Spreading and despreading in DS-CDMA.

153

Figure 10.5: Typical signal transmission and reception mechanism in CDMA communication. 10.4 CDMA CORRELATION RECEIVER Figure 10.6 depicts the operation of typical CDMA correlation receiver. Shown (in Figure 10.6) in the upper half (enclosed green solid part) is the desired signal that has been recovered after despreading operation (data X PN code followed by integration).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 10.6: Typical CDMA correlation receiver operation. However, another signal that has been spreaded with a different code from the desired signal (here it is termed as interfering signal) results in a signal after despreading operation (with the desired signal PN code) that approximates around zero value. Thus the effect of interfering signals is alleviated. It is to be noted that the amplitude of the own signal increases on average by a factor of 8 (spreading factor) relative to that of the user of the other interfering system, in term, this effect is called processing gain (the properties of the CDMA receiver that provides robustness against self-interference: a fundamental requirement for reuse factor unity. Hence, spreading/dispreading transformation does not provide any signal enhancement and, the signal improvement costs increase in transmission bandwidth 10.5 CDMA RAKE RECEIVER Multi-path propagation makes it necessary to use multiple correlation receivers so that signal energy along multi-paths can be recovered in order to recover the energy from all paths. This collection of correlation receivers, termed ‘fingers’, comprising of what is called CDMA Rake receiver. Typical multi-path propagation and corresponding delay profile is shown in Figure 10.7.

Figure 10.7: Multi-path propagation and corresponding multi-path delay profile.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

154

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 10.8 shows the block diagram of a Rake receiver with three fingers and the operation is as follows. Digitised input samples are received from the RF front-end circuitry in the form of I and Q branches (i.e. in complex low-pass number format).   

Code generators and correlator perform the despreading and integration to user data symbols. The channel estimator uses the pilot symbols for estimating the channel state which will then be removed by the phase rotator from the received symbols. The delay is compensated for the difference in the arrival times of the symbols in each finger.

155

Figure 10.8: Block diagram of CDMA Rake receiver.  

The Rake combiner then sums the channel compensated symbols, thereby providing multipath diversity against fading. In addition, matched filter is used for determining and updating the current multipath delay profile of the channel. This measured and possibly averaged multipath delay profile is then used to assign the Rake fingers to the largest peaks. 10.6 PROPERTIES OF SPREADING CODES

Multiplication with the code sequence, which is of a higher bit rate, results in a much wider spectrum. The ratio of the code rate to the information bit rate is called both the spreading factor and the processing gain of the CDMA system. In IS-95, the chipping rate is 1.2288 and the spreading factor is 64. Processing gain is usually given in dBs. To distinguish the information bit rate from the code rate, we call the code rate, chipping rate. In effect, we take each data bit and convert it into k chips, which is the code sequence. We call it the chipping rate because the code sequence applied to each bit is as you can imagine it chipping the original bit into many smaller bits. For CDMA spreading code, we need a random sequence that passes certain “quality” criterion for randomness. These criterions are: 

The number of runs of 0’s and 1’s is equal. We want equal number of two 0’s and 1’s, a length of three 0’s and 1’s and four 0’s and 1’s etc. This property gives us a perfectly random sequence.



There are equal number of runs of 0’s and 1’s. This ensures that the sequence is balanced.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



The periodic autocorrelation function (ACF) is nearly two valued with peaks at 0 shift and is zero elsewhere. This allows us to encrypt the signal effectively and using the ACF peak to demodulate quicky.

Binary sequences that can meet these properties are called optimal binary sequences, or pseudo-random sequences. There are many classes of sequences that mostly meet these requirements, with m-sequences the only ones that meet all three requirements strictly. These sequences can be created using a shift-register with feedback-taps. By using a single shift-register, maximum length sequences can be created and called often by their shorter name of m-sequence, where m stands for maximum. M-sequences M-sequences are created using linear feedback registers (LFSR). Figure 10.8 shows a three register LFSR with two different tap connection arrangements. The tap connections are based on primitive polynomials on the order of the number of

156

(a): 3 stage LFSR generating m-sequence of period 7, using taps 1 and 2.

(b): Another 3 stage LFSR generating m-sequence of period 7, using taps 2 and 3. Figure 10.8: The structure of linear feedback registers (LFSR) from which m-sequences can be created. registers and unless the polynomial is irreducible, the sequence will not be a m-sequence and will not have the desired properties. Each configuration of N registers produces one sequence of length 2N. If taps are changed, a new sequence is produced of the same length. There are only a limited number of m-sequences of a particular size. The cross correlation between an m-sequences and noise is low which is very useful in filtering out noise at the receiver. The cross correlation between any two different m-sequences is also low and is useful in Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

providing both encryption and spreading. The low amount of cross-correlation is used by the receiver to discriminate among user signals generated by different m-sequences. Think of m-sequence as a code applied to each message. Each letter (bit) of the message is changed by the code sequence. The spreading quality of the sequence is an added dimensionality and benefit in CDMA systems. Gold Sequences Combining two m-sequences creates Gold codes. These codes are used in asynchronous CDMA systems. Gold sequences are an important class of sequences that allow construction of long sequences with three valued Auto Correlation Function ACFs. Gold sequences are constructed from pairs of preferred msequences by modulo-2 addition of two maximal sequences of the same length. Gold sequences are in useful in non-orthogonal CDMA. (CDMA 2000 is mostly an orthogonal CDMA system) Gold sequences have only three cross-correlation peaks, which tend to get less important as the length of the code increases. They also have a single autocorrelation peak at zero, just like ordinary PN sequences. The use of Gold sequences permits the transmission to be asynchronous. The receiver can synchronize using the auto-correlation property of the Gold sequence.

157

Figure 10.9: Generating Gold codes by combining two preferred pairs of m-sequences. More Codes IS-95 and IS-2000 use two particular codes that are really m-sequences but have special names and uses. These are called long codes and short codes. Long code The Long Codes are 242 bits (created from a LFSR of 42 registers) long and run at 1.2288 Mb/s. The time it takes to recycle this length of code at this speed is 41.2 days. It is used to both spread the signal and to encrypt it. A cyclically shifted version of the long code is generated by the cell phone during call setup. The shift is called the Long Code Mask and is unique to each phone call. CDMA networks have a security protocol called CAVE that requires a 64-bit authentication key, called A-key and the unique ESN (Electronic Serial Number, assigned to mobile based on the phone number). The network uses both of these to create a random number that is then used to create a mask for the long code used to encrypt and spread each phone call. This number, the long code mask is not fixed but changes each time a connection is created.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

There is a Public long code and a Private long code. The Public long code is used by the mobile to communicate with the base during the call setup phase. The private long code is one generated for each call then abandoned after the call is completed. Short code The short code used in CDMA system is based on a m-sequence (created from a LFSR of 15 registers) of length 215-1 = 32,767 codes. These codes are used for synchronization in the forward and reverse links and for cell/base station identification in the forward link. The short code repeats every 26.666 milliseconds. The sequences repeat exactly 75 times in every 2 seconds. We want this sequence to be fairly short because during call setup, the mobile is looking for a short code and needs to be able find it fairly quickly. Two seconds is the maximum time that a mobile will need to find a base station, if one is present because in 2 seconds the mobile has checked each of the allowed base stations in its database against the network signal it is receiving. Each base station is assigned one of these codes. Since short code is only one sequence, how do we assign it to all the stations? We cyclically shift it. Each station gets the same sequence but it is shifted. From properties of the msequences, the shifted version of a m-sequences has a very small cross correlation and so each shifted code is an independent code. For CDMA this shift is 512 chips for each adjacent station. Different cells and cell sectors all use the same short code, but use different phases or shifts, which is how the mobile differentiates one base station from another. The phase shift is known as the PN Offset. The moment when the Short code wraps around and begins again is called a PN Roll. If I call the word “please” a short code, then I can assign, “leasep” to one user, “easepl” to another and so on. The shift by one letter would be my PN Offset. So if I say your ID is 3, then you would use the code “aseple”. A mobile is assigned a short code PN offset by the base station to which it is transmitting. The mobile adds the short code at the specified PN offset to its traffic message, so that the base station in the region knows that the particular message is meant for it and not to the adjacent base station. This is essentially the way the primary base station is identified in a phone call. The base station maintains a list of nearby base stations and during handoff, the mobile is notified of the change in the short code. There are actually two short codes per base station. One for each I and Q channels to beused in the quadrature spreading and despreading of CDMA signals. Walsh Codes In addition to the above two codes, another special code, called Walsh is also used in CDMA. Walsh codes do not have the properties of m-sequences regarding cross correlation.. IS-95 uses 64 Walsh codes and these allow the creation of 64 channels from the base station. In other words, a base station can talk to a maximum of 64 (this number is actually only 54 because some codes are used for pilot and synch channels) mobiles at the same time. CDMA 2000 used 256 of these codes. Walsh codes are created out of Haddamard matrices and Transform. Haddamard is the matrix type from which Walsh created these codes. Walsh codes have just one outstanding quality. In a family of Walsh codes, all codes are orthogonal to each other and are used to create channelization within the 1.25 MHz band. Here are first four Hadamard matrices. The code length is the size of the matrix. Each row is one Walsh code of size N. The first matrix gives us two codes; 00, 01. The second matrix gives: 0000, 0101, 0011, 0110 and so on.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

158

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

159

In general each higher level of Hadamard matrix is generated from the previous by the Hadamard transform.

Their main purpose of Walsh codes in CDMA is to provide orthogonality among all the users in a cell. Each user traffic channel is assigned a different Walsh code by the base station. IS-95 has capability to use 64 codes, whereas CDMA 2000 can use up to 256 such codes. Walsh code 0 (which is itself all 0s) is reserved for pilot channels, 1 to 7 for synch and paging channels and rest for traffic channels. They are also used to create an orthogonal modulation on the forward link and are used for modulation and spreading on the reverse channel. Orthogonal means that cross correlation between Walsh codes is zero when aligned. However, the auto-correlation of Walsh-Hadamard codewords does not have good characteristics. It can have more than one peak and this makes it difficult for the receiver to detect the beginning of the codeword without an external synchronization. The partial sequence cross correlation can also be non-zero and un-synchronized users can interfere with each other particularly as the multipath environment will differentially delay the sequences. This is why Walsh-Hadamard codes are only used in synchronous CDMA and only by the base station which can maintain orthogonality between signals for its users.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 10:10: Relationship codes used in CDMA. The above is simplified look at the use of these codes. Assume there are three users in one cell. Each is trying to talk to someone else. User 1 wants to talk to someone who is outside its cell and is in cell 2. User 3 wants to talk to someone in cell 3. Let’s take User 1. Its data is first covered by a channel Wash code, which is any Walsh code from 8 to 63. It is assigned to the user by the base station 1 in whose cell the mobile is located. The Base Station has also assigned different Walsh codes to users 2 and 3. All three of these are different are assigned by base station 1 and are orthogonal to each other. This keeps the data apart at the base station. Now based on the random number assigned by the BS, the mobile generates a long code mask (which is just the starting point of the long code sequence and is a scalar number). It now multiplies the signal by this long code starting at the mask ID. Now it multiplies it by the short code of the base station to whom it is directing the signal. When the base station receives this signal, it can read the long code and see that the message needs to be routed to base station 2. So it strips off 1st short code and adds on the short code of base station 2 which is then broadcast by the BS 1 to BS 2 or sent by landlines. BS2 then broadcasts this signal along to all mobiles in its cell. The users who is located in this cell, now does the reverse. It multiplies the signal by the BS 2 short code (it knows nothing about BS 1 where the message generated) then it multiplies the signal by the same long code as the generating mobile. How? During the call paging, the mobile was given the same random number from which it creates the same long code mask. After that it multiplies it by the Walsh code sequence (also relayed during call setup). So that’s about it with some additional bells and whistles, which we shall get to shortly. 10.7 CHANNEL WAVEFORM PROPERTIES The communications between the mobile and the base station takes place using specific channels. Figure below shows the architecture of these channels. The forward channel (from base station to mobile) is made up of the following channels:    

Pilot channel (always uses Walsh code W0) (Beacon Signals) Paging channel(s) (use Walsh codes W1-W7) Sync channel (always uses Walsh code W32) Traffic channels (use Walsh codes W8-W31 and W33-W63)

The reverse channel (from mobile to base station) is made up of the following channels: 

Access channel

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

160

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects



Traffic channel

Forward Channel Description A base station can communicate on up to 64 channels. It has one pilot signal, one synch channel and 8 paging channels. The remaining is used for traffic with individual mobiles.

Figure 10.11: Forward channels. 161

Pilot Channel Let’s start with how the base station establishes contact with the mobiles within its cell. It is continually transmitting an all zero signal, which is covered by a Walsh code 0, a all 0’s code. So what we have here is a one very long bit of all zeros. For this reason, the pilot channel has very good SNR making it easy for mobiles to find it. This all zero signal is then multiplied by the base stations’ short code, which if you recall is the same short code that all base station use, but each with different PN offset. Pilot PN Offsets are always assigned to stations in multiples of 64 chips, giving a total of 512 possible assignments. The 9-bit number that identifies the pilot phase assignment is called the Pilot Offset. This signal is real so it only goes out on the I channel, and is up-converted to the carrier frequency which in the US is 845 MHz. On the receive side, the mobile picks up this signal and notes the base station that is transmitting it. Here is a question, if the short code is cyclical, how does the receiver know what the phase offset is. Do not all the signals from all the other nearby base stations look the same? Yes, and the mobile at this point does not know which base station it is talking to, only that it has found the network. To determine of all the possible base station and there can 256 of them, each using a 512 chip shifted short code, the network uses the GPS signal and timing. The zero offset base station aligns its pilot transmission with every even second time tick of GPS. So let’s say that your mobile is in the cell belonging to a base station with PN offset ID of 10. That means that is will start its transmission 10 x 512 chip = 5120 chips after every even second time tick. So when the mobile wakes up and looks at it time, it knows exactly where each base station short code should be.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

162

Figure 10.12: Forward channel is the transmission of all traffic from the base station within its cell. All data is sent simultaneously. Then all it has to do is to do a correlation of the bits it is seeing with each of the 256 possible sequences. Of course, it tries the base station where it was last but if it has been moved then theoretical it will have to go through all 256 correlations to figure out where it is. But it does do it and at the end of the process, it knows exactly which of the base stations it is hearing. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure 10:13: The mobile looks for the code that aligns with GPS timing. It picks off the code received at this time, does a correlation with stored data and knows which base station it has found. Synch Channel The Synch channel information includes the pilot offset of the pilot the mobile has acquired. This information allows the mobile to know where to search for the pilots in the neighbor list. It also includes system time, the time of day, based on Global Positioning Satellite (GPS) time. The system time is used to synchronize system functions. For instance, the PN generators on the reverse link use zero offset relative to the even numbered seconds in GPS time. However, the mobiles only know system time at the base stations plus an uncertainty due to the propagation delay from its base station to the mobile's location. The state of the long code generator at system time is also sent to the mobile in the Synchronization message. This allows the mobile to initialize and run its long code generator very closely in time synchronism with the long code generators in the base stations. The Synchronization message also notifies the mobile of the paging channel data rate, which may be either 4800 or 9600 bits/sec. The data rate of this channel is always 1200 bps. Paging Channel Now the mobile flashes the name of the network on its screen and is ready to receive and make calls. Your paging channel may now be full of data. It may include a ring tone or a “voicemail received” message. The data on the paging channel sent by the base station, includes mobile Electronic Serial Identification Number (ESIN), and is covered by a long code. How does the mobile figure out what this long code is? At the paging level, the system uses a public long code. This is because it is not talking to a specific mobile; it is paging and needs to reach all mobiles. When the correct mobile responds, a new private long code will be assigned at that time before the call will be connected. The mobile while scanning the paging channel recognizes its phone number and responds by ringing. When you pick up the call, an access message goes back to the base station. The mobile using Qualcomm CDMA generates a 18-bit code. The mobile sends this authentication sequence to the base station during the sync part of the messaging protocol. The base station checks the authentication code before allowing call setup. It then issues a random number to the mobile, which the mobile uses in the CAVE algorithm to generate a call specific long code mask. At the same time, the base station will also do exactly that. The two now have the same long code with which to cover the messages. Traffic Channel Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

163

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The base station can transmit traffic data to as many as 54 mobiles at the same time. It keeps these channels separate by using Walsh codes. This is a code division multiplexing rather than a frequency based channelization. Walsh codes are used only by the base station and in this fashion; it is a synchronous CDMA on the forward link, whereas on the return link it is asynchronous CDMA, because there is no attempted separation between the various users. But the use of m-sequences for spreading, the quality of orthogonality although not perfect is very good. The traffic channel construct starts with baseband data at 4.8 kbps. It is then convolutionally encoded at rate of ½, so the data rate now doubles to 9.6 kbps. Symbol repetition is used to get the data rate up to 19.2 kbps. All information rates are submultiples of this rate. Data is then interleaved. The interleaving does not change the data rate, only that the bits are reordered to provide protection against burst errors. Now at this point, we multiply the resulting data sequences with the long code, which starts at the point determined by the private random number generated by both the base station and the mobile jointly. This start point is call-based and changes every time. Mobiles do not have a fixed long code assigned to them. Reverse CDMA Channel can have up to 242-1 logical channels or the total numbers of calls that can be served are 17179869184. Now the data is multiplied by a specific Walsh codes which is the nth call that the base station is involved in. Mobile already knows this number from the paging channel. The base station then combines all its traffic channels (each covered by a different Walsh code) and all paging channels (just 8) and the one pilot channel and one synch channel adds them up, does serial to parallel conversion to I and Q channels. Each is then covered by an I and a Q short code and is QPSK modulated up to carrier frequencies and then transmitted in the cell. Reverse Channel

164

In IS-95, there are just two channels on which the mobile transmits, and even that never simultaneously. It is either on the access channel or it is transmitting traffic. The channel structure is similar but simpler to the forward channel, with the addition of 64-ary modulation.

Figure 10.14: Reverse Channel from mobile to base station communication. 64-ary Modulation Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

This block takes a group of six incoming bits (which makes 26 = 64 different bit sequences of 6 bits) and assigns a particular Walsh code to each. We know that each Walsh code sequence is orthogonal to all the others so in this way, a form of spreading has been forced on the arbitrarily created symbols of 6 bits. And this spreading also forces the symbols to be orthogonal. It is not really a modulation but is more of a spreading function because we still have not up converted this signal to the carrier frequency. After this, a randomization function is employed to make sure we do not get too many 0’s or 1’s in a row. This is because certain Walsh codes have a lot of consecutive 0’s. Next comes multiplication with the long code starting at a particular private start point. Then comes serial to parallel conversion, and application of baseband filtering which can be a Gaussian or a root cosine shaping. Then the Q channel (or I, it makes no difference) is delayed by half a symbol, as shown below. The reason this is done is to turn this into an offset QPSK modulated signal. The offset modulated signal has a lower non-linearity susceptibility and is better suitable to being transmitted by a class C amplifier such as may be used in a CDMA cell phone. From there, each I and Q channel is multiplied by the rf carrier, (a sine and a cosine of frequency fc) and off the signal goes to the base station. On the demodulation side, the most notable item is the Rake receiver. Due to the presence of multipath, Rake receivers which allow maximal combining of delayed and attenuated signal, make the whole thing work within reasonable power requirements. Without Rake receivers, your cell phone would not be as small as it is. 10.8 CDMA SYSTEM ASPECTS Near/Far Problem An important problem of CDMA is the near/far problem, in which the uplink signals from distant transmitters may be overwhelmed at the receiver of the base station by much stronger signals from nearby transmitters, which are not sufficiently attenuated by the processing gain of the CDMA system. The dynamic range of path loss and shadowing at different locations can easily be 60 dB or even more. When we also include fast Rayleigh fading with dynamic range of about 40 dB, the total dynamic range of received signals can be more than 100 dB. For a CDMA system to operate successfully as a multiple access technique, the average received powers of all accessing signals must be almost equal (within 1 dB from each other). The solution to this problem is the application of very fast and very accurate power control, whereby the base station controls the transmission power of the mobile stations in such a way that the received level from all uplink transmitters at the receiver of the base station is the same. In TDMA systems power control is a feature that has been added to increase system capacity by reducing average interference level; for CDMA fast power control is essential for the successful operation of the basic concept. Considering all user‟s received power at the BS is equal, neglecting noise, the SIR can be found as follows. Eb

R

N 0  I0 B

Pr

 BN

0

 Pr ( M  1 )



1

.

M

where Eb is the per bit energy, Pr is per user received power and N0 is the noise power spectral density. Eb N0



B 1

.

R M

Thus, the Eb/N0 requirement is same for signals from all MSs, and it is determined only by the number of users. However, if the received powers from all MSs are not equal, Eb/N0 requirement is varying depending on the location of a MS within the cell. The receiver of the BS must naturally be able to receive signal from any MS, which means that the Eb/N0 requirement at the BS receiver must be chosen according to the worst case MS. When the received powers from MSs can vary over several tens of dBs without power control, the Eb/N0 requirement at the BS receiver would be also several tens of dBs higher than with power control. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

165

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

In multipath propagation environment the power control must be very fast, and it must also have a very large dynamic range. For example, in the IS-95 system the transmitter power must be adjusted 800 times in a second, which can be compared to the maximum rate of 17 power control commands within a second in the GSM system. The implementation of reliable, fast power control over very large range of power levels has been one of the most difficult design challenges for the design of CDMA systems. It should be noted that the received powers from all MSs of one cell can be controlled to be approximately equal, but the received interference from MSs of surrounding cells will still be randomly varying (according to independent power control of these cells). The near/far problem can be combated by power control within a network of a single operator. However, this problem would cause very serious problems in the case; where two operators had two independent networks using the same frequency band. This kind of scenario is quite common for FDMA and TDMA systems, but there naturally the two networks only use the same overall frequency band, but never the same frequency channels within this band. Also the FDMA and TDMA networks are suffering from the near/far problem. However, here the interference is always adjacent channel interference from one of the neighboring channels, and it can be controlled by filtering. On the other hand, the interference from other users in a CDMA system is cochannel interference occurring at the same frequency band, and it cannot be removed by filtering techniques. It should be also noted that the range of power control limits the maximum cell size in CDMA systems. Soft Handover Since the same carrier frequency is used in every CDMA cell, the receiver of each MS can receive downlink signals from all nearby BSs. Similarly, each BS can receive uplink signals also from MSs that are outside its cell. This capability of CDMA gives rise to a new approach to handover, referred to as soft handover, where mobile stations in transition between one cell and its neighbor transmit to and receive from both base stations the same data simultaneously. By using the rake receiver in the mobile, the signals from the two BSs can be isolated and aligned both in time and in phase to reinforce each other on the downlink. On the uplink, the base station controller (or the mobile switching center) must resolve which base station is receiving the stronger and hence better replica, and decide on its favor. Maximum ratio combining of the received multipath components takes place in the rake receiver of the mobile station, where the rake branches are adjusted to the strongest multipath components of all active links. Determination of the strongest multipath components (i.e. channel estimation) is accomplished with the aid of pilot channel that is transmitted by every base station. Since the rake receiver of a mobile station seeks the strongest multipath components in the received signal, it can gradually shift from receiving the old base station to the new base station. As the mobile is moving towards the new base station, all the branches of the rake receiver finally lock to multipath components of the new base station, and the transmission from the old base station can be finally ceased. On the uplink, the signal from a mobile station is received by two or more base stations during the soft handover. Soft handover is a macro diversity technique, where the optimum performance could be achieved by maximal ratio combining of the received signals of all base stations at a common node of the network (e.g. the base station controller). CDMA Capacity and Spectral Efficiency Capacity gain for CDMA over FDMA and TDMA has been often claimed, because the frequency reuse factor of unity can be used in CDMA systems. This claim misses the key point of cellular network planning. The problem of high capacity cellular networks is the cochannel interference. The frequency reuse factor, which is larger than unity, is the solution to this problem in FDMA and TDMA networks. The cochannel interference problem is very much present also in CDMA networks, where it is called multiple access interference (MAI). However, the frequency reuse is not an appropriate solution in CDMA. Instead, very precise power control is utilized to combat the effects of cochannel interference. In single cell case, for FDMA or TDMA, the number of users in a system is given by Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

166

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

B tot

M 

B tot



Bc

m ,

R

where m is the modulation efficiency, and R is the the transmission symbol rate. However, in multi-cell case, the whole system bandwidth is divided into a number of cells to address the cochannel interference and hence if K is the frequency reuse factor, then M 

B tot

m ,

R .K

which is subject to the adjacent channel interference. In CDMA, considering the co-channel interference, and assuming all user‟s received power at the BS is equal, neglecting noise, the SIR can be found as follows. Eb

SIR 

R

N 0 mB

 tot

Pr Pr ( M  1 )



1

.

M

where Eb is the per bit energy, Pr is per user received power and N0 os the noise power spectral density. M 

B tot

m

R

1

.

Eb N0

167

Note that CDMA capacity is reduced by the factor

Eb N0

which typically varies about 3 dB to 9 dB.

Hence, the maximum number of users in CDMA is given by M

SIR 

Eb



max

R

N 0  I 0 B tot

B tot R

m

1  Eb    N 0   min

Pr



N 0 B  Pr ( M  1 )



.

1 N 0B Pr

 M 1

N 0B

Pr 

B tot   Eb  R    N 0  I0 

 M  1

N 0B

 M

m ax

1 M

Hence as M approaches Mmax +1, Pr becomes infinite, which is impractical. In practical design, 50% to 75% of the theoretical maximum number of users can be used. For universal frequency reuse nalysis, considering only the first, second, and third tiers, the total interference can be found as follows.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

I   M  1  Pr  6 MP r    1 . 78  M  1  Pr ,   1 . 42  M  1  P , r

 3

n

3

n



 2 3



n

 

n  4 n 5

Hence, in multi-cell case, the total capacity is reduced from the that of the single cell scenario and is this effect is incorporated using the term frequency reuse efficiency factor F so that the CDMA system capacity becomes M 

B tot

m

R

1

F.

Eb

168

N0 1 . 78

where F  

1 . 42

1

 0 . 56 ,

n  4

1

 0 . 70 ,

n 5

. However, usually for the purpose of analysis, F=0.6 is considered.

The capacity of a CDMA system can be increased by using techniques such as discontinuous transmission with voice activity detection and cell sectorization. For 40% speech activity, theoretically improvement factor of 2.5 could be achieved from discontinuous transmission. Sectorization with 3-sector cells could provide improvement factor of 3 with ideal directional antennas or approximately a factor of 2.55 with realizable antennas. However, it should be noted that these same capacity-enhancing techniques can be used also in TDMA (or FDMA) systems. Incorporating these effects, Qualcomm Company claims the IS-95 system capacity as M 

B tot

1

R Eb

1

FG .

d N0

where d is the speech activity factor, G is the sectorization gain because of decrease in interference, F is the frequency reuse efficiency factor, and m is considered unity. 10.9 PRACTICAL DIFFERENCES BETWEEN CDMA AND FDMA/TDMA VERSATILITY WITH NEW SERVICES A CDMA system can quite easily be designed to handle speech and data services with different and/or variable bit rates, which is an important feature for multimedia-type services. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Sensitivity to Multipath Propagation Since the high chip rates of CDMA systems allow the distinct multipath components to be identified, CDMA systems can combat the effects of multipath more effectively than FDMA or TDMA systems. However, this requires complicated rake receivers to be used also in mobile stations, which may increase their cost, size and power consumption.

Sensitivity to Narrowband Interference Due to despreading and subsequent filtering in the receiver, narrowband external interference is significantly attenuated in a CDMA receiver, which makes the CDMA system more tolerant to external interference than either FDMA or CDMA systems.

Frequency Planning The complicated frequency reuse planning of FDMA and TDMA systems can be avoided in CDMA systems, which simplifies the design and operation of the cellular network.

Power Control The main disadvantages of CDMA systems are related to the power control, which is difficult to implement in practice, and which requires special coordination between base stations.

Code Allocation The allocation of spreading codes with good cross correlation properties to a large number of users in a CDMA system can be an obstacle. This problem may be especially severe in multioperator environment, where the mobile stations cannot be controlled by a single operator.

Coexistence CDMA can in some cases be introduced as an overlay to existing narrowband services, which may be very useful property for efficient usage of radio spectrum. Nevertheless, the interference between the two systems has to be carefully considered.

Faulty Equipment An important aspect of CDMA systems is that an error in the power control subsystem of the mobile station, which causes the mobile station to use too large power, causes interference to a large number of users in the serving cell as well as in the neighboring cells. Thus, a control of faulty equipment may be more critical in CDMA systems than in FDMA or TDMA systems. 10.10 CDMA Power Control Since users in the IS-95 system use the same bandwidth at the same time, transmitted signals from all other mobile stations interfere a signal from any particular mobile station. For the IS-95system to work properly, RF power in the system must be controlled, so that the received signal level from all mobile stations should be within 1 dB difference from each other. In order to minimize interference, the transmitted signal should be low as possible but still provide reliable communication. Thus, some extra functions on both mobile station and base station are required for controlling RF power. There are three major RF power control mechanism in the CDMA system. Open Loop Uplink Power Control The power level of received downlink pilot channel enables the mobile station to estimate the path loss (including shadowing) between base station and mobile station. Knowing this, the mobile station adjusts its transmitted power so that the base station receives its signal at approximately constant level. The mobile station starts to control power by estimating the received power from the downlink channel. This power estimate then contains the total power at the input of the receiver comprising the desired signal as well as noise and interference. Mobile station employs the received power to adjust its transmit power. When received power decreases, the transmit power is increased and vice versa. The open loop power control takes care of most of the vast Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

169

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

dynamic range of the path loss; fine tuning of the output power is then achieved by the closed loop power control. The open loop power control also determines the initial transmission power of the MS. Closed Loop Uplink Power Control Since in the open loop power control the mobile station estimate path loss from a downlink channel which is at a 45 MHz higher frequency than the uplink channel, the open loop power control cannot follow the fast fading characteristics of the uplink channel. The path loss and shadowing on uplink and downlink bands are very nearly reciprocal, but the fast fading phenomena on these two different frequency bands are different and the instantaneous attenuations of the received signals are also different. To make power control to follow fast fading, the base station measures the received signal-tointerference ratio(SIR) over a 1.25-ms period, compares that to the target SIR, and transmits power adjustments back to the mobile station at the rate one command (two bits with the same value: "0" for increasing power or "1" for decreasing power) in every 1.25 ms by puncturing the data symbols on the downlink traffic channel. The location of power control bits is randomized within the 1.25 ms block of data symbols. The commands are always for either increasing or decreasing power, so that maintaining of a constant power level requires transmitting of alternating increase and decrease commands. The mobile station extracts the power control bits from the received signal and adjusts its transmission power accordingly. The adjustment step is a system parameter and can be 0.25, 0.5, or 1.0 dB. The dynamic range for the closed loop power control is ±24 dB. The composite dynamic range for open and closed loop power control is ±32dB. Since the SIR required to produce acceptable bit error rate varies in different multipath radio environments, the IS-95 system employs an outer loop that adjusts the target SIR. The base station measures the signal quality (bit error rate) from the received data, and based on that determines the target SIR. • The rate of 800 commands per second is high enough to follow fast fading for practical velocities of MSs. Since the power control commands are transmitted without any type of channel coding, their bit error ratio can be fairly high (on the order of 5%). However, since the power is only adjusted by one step up or down after each command, this high BER does not cause any serious problems. During a soft handover, each participating base station makes its own power control decision, independent of the others, unless they are different sectors of the same cell, in which case they all transmit a common decision. To minimize the interference in a CDMA system, during a soft handover the power level of a MS should be controlled by the BS which is receiving the best-quality signal. To implement this, a MS is monitoring downlink power control bits from all BSs participating in the handover, which means that if any of the base stations commands for power decrease, then the mobile station is required to reduce its uplink power. Except for this feature of the power control, during a soft handover a mobile station is transmitting exactly the same way that it would in the absence of the handover. Downlink Power Control In downlink, signals for all mobiles propagate through the same multipath channel from the serving BS to any MS and thus are received by a mobile station with equal attenuation. Therefore, no power control is required to eliminate the near-far problem, and the downlink power control is less critical to the overall system performance than the uplink power control. However, it should be noted that the interference from surrounding cells still exhibit fading that is uncorrelated to the fading characteristics of the desired signal. Thus, downlink power control is still necessary to minimize the interference to other cells and to compensate against the interference from other cells. In order to keep the total RF power on the downlink channel low, to maximize the number of mobile stations and to avoid disturbing neighboring cells, the base station always tries to send RF power as low as possible. The base station periodically reduces its transmitting power in small steps until one of the communicating mobile stations encounters too high BER and requests for higher RF power from the base station. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

170

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

This adjustment occurs every 15 ms to 20 ms. The dynamic range of the downlink power control isonly ±6 dB. In IS-95 system, it is often assumed that only one carrier frequency is used in the whole CDMA network. However, when capacity demand is high enough, it is naturally necessary to allocate multiple carriers in a single cell. Another reason for using more than one carrier is hierarchical network design, where a macrocell overlaying microcells uses different carrier than the microcells. Since CDMA transmission is continuous, there are no idle slots for the inter-frequency measurement as in TDMA-based systems. For that reason, the IS-95 system cannot support signal quality measurements at different carriers, and inter-frequency handovers have to be executed as blind handovers without knowledge of the signal quality at the new carrier. Mathematical Analysis Assume that there is only one user of the system. The carrier power C  SNR 

Eb

 R Eb

Tb

If we define the transmit power equal to W and signal bandwidth equal to B, then the interference power at the receiver is equal to I  WN

0

Now we can write Eb C I



R Eb

N0



171

W

W N0

R

The quantity W/R is the processing gain of the system. Now let’s call M the number of users in this system. The total interference power is equal to I  C M  1

Substituting this in the above equation, we get, C



I

C C M  1

1



M 1

and with one more substitution we get Eb C I



N0 W



R M  M 1

1 M 1 W

1

R Eb N0

So the system capacity is a direct function of the processing gain for a given Eb/N0. Note that we made an assumption that all users have similar power level so the interferences are additive. No one user overwhelms all the others. If the power levels of all users are not equal then the system capacity is compromised and the C/I expression above is not valid. Hence, the CDMA systems manage the power levels of all mobiles so that the power level of each mobile is below a certain required level and is about the same whether the mobile is very close to the base station Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

or far at the edge of the cell. Multipath and fading also attenuate power levels so the system maintains a power control loop. The above equation does not take into account the background thermal noise,  in the spread bandwidth. To take this noise into consideration, Eb/N0 can be represented as Eb

W 

N0

R

M

   1      C

W

M 1

     Eb  C    N 0   R

Note that the background noise determines the cell radius for a given transmitter power. In order to achieve an increase in capacity, the interference due to other users should be reduced. The first technique for reducing interference is antenna sectorization. As an example, a cell site with three antennas, each having a beam width of 1200, has interference, which is one-third of the interference received by an onini-directional antenna. This increases the capacity by a factor of 3 since three times as many users may now be served within a sector while matching the performance of the omni-directional antenna system. Looking at it another way, the same number of users in an omni-directional cell may now be served in 1/3 the area. The second technique involves the monitoring of voice activity such that each transmitter is switched off during periods of no voice activity. Voice activity is denoted by a factor  , and the interference becomes (MS-1)  , where MS is the number of users per sector. With the use of these two techniques, the new average value of

Eb

S

within a sector is given as

N0

Eb

W

S



N0

R

M S

   1       C

When the number of users is large and the system is interference limited rather than noise limited, the number of users can be shown to be

M

S

 W 1  R 1 S   Eb   N0

     

Note that if the voice activity factor is assumed to have a value of 3/8, and three sectors per cell site are used, the above equation demonstrates that the SNR increases by a factor of 8, which leads to an 8 fold increase in the number of users compared to an omni-directional antenna system with no voice activity detection. Call Origination

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

172

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

When the user decides to make a call, the MS uses the uplink access channel to contact the serving BS. As link is yet established, closed loop power control cannot be used, and the MS can use only open loop power control to adjust its output power. Since collisions of transmission from several MSs can take place on the access channel, a MS may require multiple access attempts, which are separated by random time intervals. After each attempt, the MS listens to the downlink paging channel for response from the BS. After successful access, the BS responds with a traffic channel assignment message. After this the MS can switch to the traffic channel, and call conversation can start. Authentication Authentication in the IS-95 system closely resembles authentication in the GSM system. During authentication in both systems:  network creates a random number R and sends it to a MS,  MS computes a response S from R by using authentication algorithm and sends it to network,  network computes locally a response S' from R by using the same authentication algorithm as in MS and compares the result to the response S received from the MS.  Only when S = S' authentication is accepted. However, there are subtle differences in the details of IS-95 authentication process. 10.11 CDMA 2000 This is an evolution and extension of capabilities and builds on the IS-95 standard. One of the big ways in which CDMA 2000 differs from IS-95 is that it includes beam forming. Each base station cell is now divided in three sectors such that frequency is reused. This increases the gain at the mobile and allows better SNR and a larger number of users. The other significant way that IS-2000 differs from IS-95 is that it allows additional forward and reverse channels. Some of these channels are the same as IS-95 and others are new. Spreading codes are also changed to allow larger data rates. The 1.25 MHz channel with the 1.2288 mbps spreading rate called 1X can now be 3X 93 x 1.2288 mbs) or 5X (5 x 1.2288 mbps). 10.12 SUMMARY In CDMA systems, the narrowband message signal is multiplied by a very large bandwidth signal called the spreading signal. The spreading signal is a pseudo-noise code sequence that has a chip rate which is orders of magnitudes greater than the data rate of the message. All users in a CDMA system use the same carrier frequency and may transmit simultaneously. Each user has its own pseudorandom codeword which is approximately orthogonal to all other codewords. The receiver performs a time correlation operation to detect only the specific desired codeword. All other codewords appear as noise due to decorrelation. For detection of the message signal, the receiver needs to know the codeword used by the transmitter. Each user operates independently with rio knowledge of the other users. In CDMA, the power of multiple users at a receiver determines the noise floor after decorrelation. If the power of each user within a cell is not controlled such that they do not appear equal at the base station receiver, then the near-far problem occurs. The near-far problem occurs when many mobile users share the same channel. In general, the strongest received mobile signal will capture the demodulator at a base station. In CDMA, stronger received signal levels raise the noise floor at the base station demodulators for the weaker signals, thereby decreasing the probability that weaker signals will be received. lb combat the near-far problem, power control is used in most CDMA implementations. Power control is provided by each base station in a cellular system and assures that each mobile within the base station coverage area provides the same signal level to the base station receiver. This solves the problem of a nearby subscriber overpowering the base station receiver and drowning out the signals of far away subscribers. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

173

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Power control is implemented at the base station by rapidly sampling the radio signal strength indicator (RSSI) levels of each mobile and then sending a power change command over the forward radio link. Despite the use of power control within each cell, out-of-cell mobiles provide interference which is not under the control of the receiving base station. The features of CDMA including the following:  Many users of a CDMA system share the same frequency. Either TDD or FDD may be used.  Unlike TDMA or FDMA, CDMA has a soft capacity limit. Increasing the number of users in a CIJMA system raises the noise floor in a linear manner. Thus, there is no absolute limit on the number of users in CDMA. Rather, the system performance gradually degrades for all users as the number of users is increased, and improves as the number of users is decreased.  Multipath fading may be substantially reduced because the signal is spread over a large spectrum. If the spread spectrum bandwidth is greater than the coherence bandwidth of the channel, the inherent frequency diversity will mitigate the effects of small-scale fading.  Channel data rates are very high in CDMA systems. Consequently, the symbol (chip) duration is very short and usually much less than the channel delay spread. Since PN sequences have low autocorrelation, multipath which is delayed by more than a chip will appear as noise. A RAKE receiver can be used to improve reception by collecting time delayed versions of the required signal.  Since CDMA uses co-channel cells, it can use macroscopic spatial diversity to provide soft handoff. Soft handoff is performed by the MSC, which can simultaneously monitor a particular user from two or more base stations. The MSC may choose the best version of the signal at any time without switching frequencies.  Self-jamming is a problem in CDMA system. Self-jamming arises from the fact that the spreading sequences of different users are not exactly orthogonal, hence in the despreading of a particular PN code, non-zero contributions to the receiver decision statistic for a desired user arise from the transmissions of other users in the system.  The near-far problem occurs at a CDMA receiver if an undesired user has a high detected power as compared to the desired user.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

174

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

EXERCISE Exercise 01 A Base Station antenna is installed at the height of 25 meters in Dhaka University area and radiates the signal at the carrier frequency 1800 MHz. A mobile station is 2 km away from BS and its height is 3 meters. If the receiver sensitivity is ‐60 dB, the link between the MS and BS will be established or not. If the link is not working, how can you establish the link? Explain in details. Note that the transmit power is 2 watts (maximum). Exercise 02 Determine the max. and min. spectral frequency received from a stationary 1800/1900 GSM transmitter that has a centre frequency of exactly 1960 MHz, Assuming that a MS is travelling at the speed of 1, 5, 100Km/60 min. Exercise 03 Plot the path loss for urban, suburban and rural areas as a function of distance for ranges from 2‐20 km (Hata Model). The height of the BS and MS are 30 m and 2 m respectively. The carrier frequency is 900MHz. Exercise 04 Assume a receiver is located 10 km from the 50W transmitter. The carrier frequency is 6 GHz and free space propagation is assumed. Gt=Gr=1. Find the power at the receiver, the rms voltage applied to the receiver input , assume the receiver antenna has a real impedance of 50 ohm and is matched to the receiver Exercise 05 A single branch Rayleigh fading signal has a 20% change of being 9 dB below some mean SNR threshold. Determine the mean of the Rayleigh fading signal as referenced to the threshold. Find the likelihood that a four branch selection diversity receiver will be 9 dB below the mean SNR threshold. Exercise 06

Signal to co-channel interference ratio estimation Now consider the signal to co-channel interference ratio as given below where the reuse factor K equals to 7. C I

R



4

6D

4

Find the signal to co-channel interference ratio in the design of cellular network (as given in Figure 4.1) for the following cases: (a) Case 1: when only first-tier is considered. (b) Case 2: when also the second tier of next closest co-channels cells is included in the interference power. You can approximate the distance between the mobile station and a second-tier interferer as the distance between the center of the serving cell and the center of the interfering cell. (c) Case 3: when also the third tier of the next closet co-channel interferers is included in the interference power.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

175

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

3rd tier

A Co-channel cell

A

2nd tier

Co-channel cell

A Co-channel cell

A

1st tier

Co-channel cell

D

A R Serving cell

Figure 4.1: Hexagonal cellular network with reuse facto K=7.

176

Exercise 07

Power delay profile (a) The power delay profile of indoor and outdoor is given as follows. Time  Ph(  )

0 0

Time  Ph(  )

0 -15

Indoor 25 -10 Outdoor 25 -20

50 -15

ns dB

50 -5

ns dB

Determine whether or not the GSM system in both environments shows channel frequency selective fading property. (b) Now consider that a MS is moving with 150 km/hr speed in open areas. (i) Find the maximum and minimum spectral frequencies in the received signal. (ii) Determine whether or not the channel shows time selective fading if the symbol rate is considered 10 ksymbol/s. (iii) Find the maximum number of symbols that could be transmitted without updating the equalizer. Exercise 08

Consider a mobile communication system where a base station transmits signal to the mobile station as shown in figure 1. Because of the multi-path propagation effect, the signal received at the mobile station experiences delay (in ns) differently along different paths, and the signal strength in terms power (in dB) along each path is shown in the figure. The system was designed as such that the maximum speed of the mobile station would be 60 km/hr.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

(a) As a system designer, determine the channel bandwidth and the duration of the transmitted burst so that the channel shows flat response to both frequency and at least over one burst time duration. You can consider any reasonable carrier frequency for your design consideration. (b) If the total system bandwidth is 50 MHz, how many such channels can be allocated? You can consider a near approximate value since channels cannot be fractionalized. { P(dB), delay (nS)}

{-5 ,

0.0

33

}

{0, 0}

{-10,

0.05

5} Mobile Station

,0 {-1

.1}

177

Base Station

Figure 1: A typical multi-path propagation in cellular mobile communication systems. Exercise 09

Hand-off operation A base station (BS) estimates the uplink signal from a mobile station (MS) which is 4 km away from the BS, and the estimated average received signal level is -80 dBm. Consider that the MS is moving directly away from the BS at the velocity of 50 meters/second, during the time of estimation. The carrier frequency of the received signal is 900MHz, and the base station antenna height is 30m. Assume that the path loss follows Hata model and the degradation in the received signal strength is linear with distance. Determine how much time is left before a hand over must be executed when the threshold for hand over is set at the average received signal level of -100 dBm. Hints: you may consider the path loss equation in Hata model. Exercise 10

Cell sectorization (a) Consider an AMPS system, with 300 voice channels, cluster size N = 7, and omnidirectional antennas at the base stations. Base stations are located at the center of the cells and transmit the same power on the forward link. Consider the first tier of co-channel cells only. Find the minimum SIR on the forward link. (b) Now consider the forward link only for sectoring to increase SIR. Two sectorized antennas are available: beamwidth (BW) = 60° for six sectors per cell, and BW = 120° for three sectors per cell. Assume that all cells have hexagonal shape, with radius R. Determine the minimum SIR at the mobile when the mobile is located at the cell boundary for three and six sectors. Note that the number of interfering base stations in the first tier 6, 2, and 1 for no sectoring, 1200, and 600 sectoring respectively. Exercise 11 Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Co-channel cell distance estimation Consider a hexagonal cellular geometry as shown in the figure 4.1 where D is the distance between cochannel cells, and R is the radius of a cell. Assume that the co-channel cell received power must be at least -120 dBm to satisfy the minimum co-channel interference requirement for link quality. All cells are centerexcited, and the received power at distance d0 =1 km is 2 mW from the BS. The path loss exponent n is 2. Note that here A in the figure denotes a co-channel cell. Consider free space pathloss model. (a) Find the value of D so that the co-channel interference requirement as aforementioned is satisfied. Hints: you may consider finding the path loss at distance, D. (b) If we consider a reuse factor of 3, 7, and 12, find the corresponding cell radius, for each value of reuse factor. Determine and justify which of these reuse factors can provide more capacity than the other? Exercise 12

Link budget estimation Consider a GSM system with the followings: Base station (BS) carrier frequency: 0.9 GHz RF power of BS transmitter: 46 dBm BS antenna gain Gt : 14 BS antenna height hb : 100 m BS cable loss: 2dB Mobile station (MS) antenna gain Gr: 1 MS antenna height is hm : 2 m BS cable loss: 0 dB Distance between BS and MS: d (km) Fading margin: 8 dB Hata model is considered for the path loss estimation

178

(a) Now consider that the path losses in both urban and open areas are the same. Find the value of d in both urban and open areas. (b) Find the path loss in both urban and open areas. (c) Also find the received RF power at the antenna front end of the MS. (d) Consider hexagonal cell shape. Find the area of a cell in both urban and open areas. Exercise 13

Business case in cellular networks A cellular network is to cover 200 square km. Assume a base station costs $400,000 and a MTSO costs $1,200,000. An extra $600,000 is needed to advertise and start the business. (a) How many base stations (i.e., cell sites) will you be able to install for $7 million? (b) Assuming the earth is flat and subscribers are uniformly distributed on the ground, determine the coverage area of each of your cell sites? What is the major radius of each of your cells, assuming a hexagonal cell shape? (c) Assume that the average customer will pay $30 per month over a five year period. Assume that on the first day you turn your system on; you have a certain number of customers which remains fixed throughout the year. On the first day of each New Year, the number of customers using your system doubles and then remains fixed for the rest of that year. What is the minimum number of customers you must have on the first day of service in order to have earned $15 million in gross billing revenues by the end of the 5th year of operation? (d) For your answer in (c), how many users per square km are needed on the first day of service in order to reach the $50 million mark after the 5th year? Exercise 14 Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

Link budget estimation A Base Station antenna is installed at the height of 35 meters in Gulshan area and radiates the signal at the carrier frequency 900 MHz. A mobile station is 1 km away from BS, and its height is 2 meters. If the receiver sensitivity is ‐80 dB, determine whether or not the link between the MS and BS will be established. If the link is not working, how can you establish the link? Explain in details. Note that the transmit power is 2.5 watts (maximum). Exercise 15

Doppler frequency Determine the maximum and minimum spectral frequency received from a stationary 1800 GSM transmitter that has a centre frequency of exactly 1960 MHz. Assume that a MS is travelling at the speed of 1 km/60 min, 5 km/60 min, and 100 km/60 minutes. Exercise 16

Receiver sensitivity Assume a receiver is located 13 km from the 40W transmitter. The carrier frequency is 6 GHz, and free space propagation is assumed. Both transmit and receive antenna gains are 0 dB. Find the power at the receiver and the rms voltage applied to the receiver input. Assume that the receiver antenna has a real impedance of 50 ohm and is matched to the receiver. Exercise 17

Traffic intensity (a) A receiver in an urban cellular radio system detects a 5 mW signal at d = d0 = 2 meter from the transmitter. In order to mitigate co-channel interference effects, it is required that the signal received at any base station receiver from another base station transmitter which operates with the same channel must be below –90 dBm. The path loss exponent in the system is n = 4. Determine the major radius of each cell if a seven-cell reuse pattern is used. (b) If the system is operated with 610 channels, 15 of which are designated as setup (control) channels so that there are about 85 voice channels available per cell. If there is a potential user density of 2000 users/km2 in the system, and each user makes an average of one call per half an hour and each call lasts 0.5 minute during peak hours, find the traffic intensity of the system. Exercise 18

System design The AMPS system is allocated 50 MHz of spectrum in the 800 MHz range and provides 832 channels. Forty-two of those channels are control channels. The forward channel frequency is exactly 45 MHz greater than the reverse channel frequency. (a) What is the bandwidth for each channel and how is it distributed between the base station and the subscriber? (b) Assume a base station transmits control information on channel 352, operating at 880.560 MHz. What is the transmission frequency of a subscriber unit transmitting on channel 352? (c) The uplink and downlink cellular carriers evenly split the AMPS channels. Find the number of voice channels and number of control channels for each carrier. Exercise 19

Business case in cellular networks A cellular network is to cover 140 square km. Assume a base station costs $500,000 and a MTSO costs $1,500,000. An extra $500,000 is needed to advertise and start the business. (a) How many base stations (i.e., cell sites) will you be able to install for $6 million? Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

179

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

(b) Assuming the earth is flat and subscribers are uniformly distributed on the ground, determine the coverage area of each of your cell sites? What is the major radius of each of your cells, assuming a hexagonal cell shape? (c) Assume that the average customer will pay $50 per month over a four year period. Assume that on the first day you turn your system on; you have a certain number of customers which remains fixed throughout the year. On the first day of each New Year, the number of customers using your system doubles and then remains fixed for the rest of that year. What is the minimum number of customers you must have on the first day of service in order to have earned $10 million in gross billing revenues by the end of the 4th year of operation? (d) For your answer in (c), how many users per square km are needed on the first day of service in order to reach the $10 million mark after the 4th year? Exercise 20 Consider a hexagonal cellular geometry with cluster size K=13. Assume shift parameters i and j where j < i.

180 R

A

(a) Find the value of i and j so that the conditions aforementioned are satisfied, i.e. K=13 and j < i. (b) Consider the cells in a cluster is denoted as 1, 2, 3, ..., 13. And the cell levelled with A is levelled as the cell 1 in a cluster. Consider that A is the desired cell of concern. (i) Find all the co-channel cells in the figure and level them with A. Note that cell A is basically cell 1 in the denotation as aforementioned. (ii) Show atleast one cluster levelled with cells using the aforementioned denotation, i.e. 1, 2, 3, ... , 13 so that the nearest cells are levelled with no successive numbers to reduce adjacent channel interference. (iii) Show the first-tier, the second-tier and, the third-tier of co-channel cells of cell A. (iv) Find the co-channel cell distance from the desired cell A for the first-tier co-channel cells only. (v) Find the value of C/I, considering only the first-tier, and that the MS is located at the cell radius distance away from the BS. Note that we consider all cells are center-excited. (c) If a GSM system is launched using this cellular pattern with system bandwidth 50 MHz, and the channel bandwidth 200 kHz, Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

(i) Find the total number of channels per cell in the system. (ii) Find the capacity of the system if the cluster is reused 5 times to cover a particular area. (d) Now assume each cell phone user generates traffic at the rate 0.5 calls/s with mean holding time 100ms. The traffic is equally distributed among the channels. Assume that if there is no channel available, the requesting user is blocked without access to the network. (i) Find the traffic intensity per channel in a cell at the busy hour (when all channels are busy). (ii) Find the carried traffic offered at the busy hour in a cell if on average one user cannot get access to the network. (iii) Find the call blocking probability at the busy hour in a cell. (e) Now consider that a lost call is delayed if it is not immediately served. (i) Find the average delay for calls which are queued only and for all calls in a cell. (ii) Also find the grade of service for the case of delays being greater than 10 seconds. Exercise 21 Consider a GSM 900 system with system bandwidth 50 MHz. Consider 26-TDMA multiframe, and a user can access one time slot out of eight in a TDMA frame to transmit TCH/F. For channel coding, 1/3 convolutional coder is used. a) Find the maximum number of slots that a user can access in a multiframe. b) Find number of information bits in a normal burst. c) If a user needs to transmit a message with 1710 bits (after channel coding), find the interleaving delay to transmit the whole message. d) Consider 1.2 MHz frequency spacing for frequency diversity. Find the range of frequency over which the message (in part (c)) _is transmitted. e) If the user at any arbitrary TTI downloads information on 910 MHz, find the corresponding channel that the user can use and the minimum delay (reception and transmission) to transmit information to the BTS. f) Consider 200 kHz on each edge is allocated for the system guard period. Find the maximum number of users that can simultaneous access the system in the downlink if 2 percentages of the total channels in the system is reserved for the downlink control channel. g) Now Consider an AMPS system with the same bandwidth in the downlink as that of GSM (in part (f)). Find the maximum number of downlink users that can access the system simultaneously. h) Develop a mathematical relation between the capacity (simultaneous maximum number of users) of AMPS and GSM systems with the same bandwidth (as found in part (f)).

Use any relevant assumptions, data, or specifications of GSM and AMPS systems with necessary justification.

Exercise 22 Answer in brief the followings. a) Why does GSM encryption algorithm use 22-bit TDMA frame number in enciphering and deciphering procedures? b) Why does the circuit-switched network insufficient for bursty data applications? c) Although in GPRS, the MSC can page the MS for incoming circuit-switched calls via the SGSN, traffic channels are normally transferred using circuit-switched networks. Explain the statement. d) What are the advantages of resegmentation of radio blocks in EDGE? e) Why does two steal flag bits use in a GSM normal burst? f) In GSM, the guard band used in an access burst is relatively much higher than in a normal burst. Justify the statement. g) For necessity, the training sequence in synchronization burst is unique although should be different in normal burst in GSM neighboring cells? Justify the statement. h) In GSM system, two special blocks are used in the transmission of TCH/FS logical channel. Name those blocks and explain their functions. i) Why does fast power control necessary in CDMA? j) CDMA is more susceptible to interference than GSM. Do you agree? Justify your reason. k) Increase in user, simultaneously served, degrades the received signal quality in CDMA. Justify the statement. l) Explain why Near-Far problem needs considerable attention in CDMA system design. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

181

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

m) Why does the training sequence and guard band use in GSM system? n) Why does the alternative TMSI to IMSI use in GSM system? o) Explain in brief the methods to increase user data rates in EDGE Evolution.

RFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26]

[27]

A. Goldsmith, 2005. “Wireless Communications”, Cambridge University Press. B. Sklar (1997, September). “Rayleigh Fading Channels in Mobile Digital Communication Systems Part I: Characterization”, IEEE Communications Magazine, pp. 136-146, September 1997.. S. Faruque, (1996).Cellular Mobilesystems Engineering. Norwood, MA 02062. Artech House, INC. D. J. Gibson, (1996).The Communications Handbook.USA. CRC Press, Inc. M. Hata, (1998). Propagation Loss Prediction Models for Land Mobile Communications. IEEE, 15 18. M. Hata, (1980). Empirical Formula for Propagation Loss in Land Mobile Radio Services. IEEE Transactions On Vehicular Technology, 29 (3), 317-325. K. M. Ahmed, 2009. “AT 77.07: Cellular Mobile Systems”. Lecture notes. Telecommunications Engineering, Asian Institute of Technology, Bangkok, Thailand. D. Parsons, (1994).The Radio Propagation Channel. London. PENTECH PRESS. S. Ranvier, (2004). Physical layer methods in wireless communication systems. (Lecture notes, Course no.S-72.333, SMARAD Centre of Excellence). Helsinki: Helsinki University of Technology. K. T, Sarkar, et al. (2003, June). A Survey of Various Propagation Models for Mobile Communication. IEEE Antennas and Propagation Magazine, 45(3), 51-82. R. S. Saunders (1999). Antenna And Propagation For Wireless Communication Systems. Chichester, England. John Willy & Sons Ltd. T. S. Rappaport, 2002. “Wireless Communications; Principles and Practice”. Prentice-Hall Inc., NJ. J. Walfisch & L. H. Bertoni, (1988). A Theoretical Model of UHF Propagation in Urban Environments. IEEE Transactions On Antennas And Propagation, 36 (12), 7788 – 7796. B. Ghribi & L. Logrippo (2000). “Understanding GPRS: the GSM packet radio service”, ELSEVIER, Computer Networks, vol. 34, pp. 763-779. H. Granbohm & J. Wiklund (1999). “GPRS-General packet radio service”, Ericsson Review, vol. 2, pp. 82-88. D. Yamaguchi, et al., “EDGE Introduction to High-Speed Data in GSM/GPRS Networks” whitepaper, microsite mobile. Håkan Axelsson, et al., (2006). “GSM/EDGE continued evolution”, Ericsson Review Vol. 1, pp. 2029. Ericsson (Sep 2009). “The evolution of EDGE”. White Paper. Rysavy Research, (September 2007). “EDGE, HSPA, LTE: The Mobile Broadband Advantage”. 3G Americas. H. Holma and A. Toskala, “WCDMA for UMTS: Radio Access for Third Generation Mobile Communications”, John Wiley & sons, Ltd., third edition, 2004. C. Langton (2002). “Intuitive Guide to Principles of Communications: CDMA Tutorial”, retrieved from www.complextoreal.com. K. Kettunen (1997). “Code Selection for CDMA Systems”, Signal Processing in Communications, FALL -97, VTT, Information Technology, FIN-02044 VTT. E. Dahlman et al. (2011), “4G LTE/LTE-Advanced for Mobile Broadband”, Academic press, UK. Chae et al. (2010, May), “Adaptive MIMO Transmission Techniques for Broadband Wireless Communication Systems”, IEEE Communications Magazine, pp. 112-118. Stüber et al. (2004, February), “Broadband MIMO-OFDM Wireless Communications”, Proceedings of the IEEE, Vol. 92, No. 2, pp. 271-293. Mietzner et al. (2009), „‟Multiple-antenna Techniques For Wireless Communications – A Comprehensive Literature Survey‟‟, IEEE Communications Surveys & Tutorial, vol. 11, No. 2, second quarter, pp. 87-105. K. M. Ahmed (2010), „‟Cellular Mobile Systems‟‟, Lecture notes on AT 77.07, Asian Institute of Technology, Bangkok, Thailand.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

182

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

[28] [29] [30] [31] [32]

[33]

[34] [35] [36]

P. Saengudomlert (April 2011), “Fundamentals of Orthogonal Frequency Division Multiplexing (OFDM)”. Telecommunications program, Asian Institute of Technology (AIT), Bangkok, Thailand. R.K. Saha, “Resource reuse and scheduling in femtocell deployed two-tier LTE-Advanced Systems”, Ait master thesis-ict, asian institute of technology, bangkok, thailand, may 2011. Wang, Y., et al. (2010). Carrier Load Balancing and Packet Scheduling for Multi-Carrier Systems. IEEE Transactions on Wireless Communications, 9 (5), 780-1789. Chen, L., et al. (2009). Analysis and Simulation for Spectrum Aggregation in LTEAdvanced System. IEEE. Retrieved May 25, 2010, from http://ieeexplore.ieee.org/. Pokhariyal, A., et al. (2007). Frequency Domain Packet Scheduling Under Fractional Load for the UTRAN LTE Downlink. IEEE Online. Retrieved June 30, 2010, from http://ieeexplore.ieee.org/. Pokhariyal, A., et al. (2007). HARQ Aware Frequency Domain Packet Scheduler with Different Degrees of Fairness for the UTRAN Long Term Evolution. IEEE Online. Retrieved June 30, 2010, from http://ieeexplore.ieee.org/. S. Parkvall, E. Englund, M. Lundevall, and J. Torsner (2006, Feb), “Evolving 3G Mobile Systems: Broadband and Broadcast Services in WCDMA”, IEEE Communications Magazine, pp. 68-74. E. Dahlman, S. Parkvall, J. Sköld, and P. Beming, “3G Evolution: HSPA and LTE for Mobile Broadband”, Second edition, Academic Press , Linacre House, Jordan Hill, Oxford, OX2 8DP, 2008. M. Schwartz (2005), “Mobile Wireless Communications”, Cambridge University Press, The Edinburgh Building, Cambridge CB2 8RU, UK.

183

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

APPENDIX01 CELLULAR RADIO HISTORY AND EVOLUTION Marconi patents the wireless telegraph (1895) • ship-to-shore radiotelephone service initiated on the east coast of USA (1919) • the first mobile radio for police at Detroit, USA (1921) • the first two-way mobile radio system for police at New Jersey, USA (1933) • frequency modulation invented by Edwin H. Armstrong (1935) • the interconnection of mobile users to the public telephone network (mobile telephone service) in St. Louis, Missouri, USA (1946). • the cellular network concept proposed in Bell Labs (1947) • the first fully automatic (switching) mobile telephone system in Richmond, Indiana, USA (1948) • the bandwidth of FM mobile radiotelephones in USA decreased from the original 120 kHz to 60 kHz in 1950, and to 30 kHz at early (1960s). • in early 1950s manual trunking, i.e. each mobile user was able to manually choose among a group of channels instead of using a fixed frequency, by early 1960s automatic trunking, i.e. the radio system automatically searches for a free channel without user intervention in USA. • direct dialing and full-duplex operation at mid (1960s) • FCC allocates the 800 MHz frequency range for cellular mobile radio in USA (1970) • High Capacity Automobile Telephone System (HCMTS) analog cellular system operation started by NTT in Japan (1980). • Nordic Mobile Telephone (NMT) analog cellular system began operation at 450 MHz frequency range in Nordic countries in Europe (1982) • Advanced Mobile Phone System (AMPS) analog cellular system began operation in Chicago, USA (1983) • Total Access Communication System (TACS, modification of AMPS) analog cellular system began operation at 900 MHz frequency range in United Kingdom (1985) • Nordic Mobile Telephone (NMT) analog cellular system was extended to 900 MHz frequency range in Nordic countries in Europe (1986). • Global System for Mobile communications (GSM) TDMA digital cellular system began operation at 900 MHz frequency range in Europe (Finland) (1991) • DCS 1800 digital cellular system (extension of GSM) began operation at 1800 MHz frequency range in Europe (1993). • Introduction of TDMA digital cellular systems in USA and Japan and a CDMA digital cellular system in USA during the first half of 1990s • NMT-900 analog cellular system was shutdown in Nordic countries, Netherlands and Switzerland in 2000 to free more bandwidth for the GSM system. • standardization work for third generation cellular systems (IMT2000 and UMTS standards) from 1991; the first commercial 3rd generation cellular system has been opened for limited use in Japan in 2001.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

184

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

185

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

APPENDIX 02 MODULATION TECHNIQUES OVERVIEW Depending on the system under study, a channel can be a frequency band, a time slot in a frequency band, or a unique code, a sequence of defined binary symbols. In each of these cases, communication is carried out using a specified carrier frequency, with various modulation techniques used to transmit the desired information, whether it be voice, data, or other types of information-bearing signals, over this carrier. In this chapter we describe various modulation techniques proposed or adopted to utilize the channels most effectively. Modulation or variation of the frequency of a carrier is universally used in communication systems to bring the information to be transmitted up to a desired operating or carrier frequency. In the case of mobile wireless systems, the resultant modulated radio signal is then transmitted uplink, from mobile to base station, or downlink, from base station to mobile, as the case may be. The transmitted power is the power in this modulated carrier signal. First-generation cellular systems, such as the AMPS system used analog FM as the modulation scheme. Second-generation systems, such as digital AMPS or D-AMPS (also referred to as IS-54/136), IS95, or GSM use digital modulation schemes. Third-generation systems and the personal communication networks use digital modulation techniques as well. Consider second-generation systems as examples. In the USA, as noted in Chapter 1, mobile cellular systems have been assigned the frequency band of 824–849 MHz for uplink communications and 869–894 MHz for downlink transmissions. Both AMPS and D-AMPS slot the 25 MHz-wide bands available in each direction into 832 30 kHz channels. D-AMPS further uses time slots in each 30 kHz band to increase the capacity of the system by a factor of 3–6. The modulation technique chosen for D-AMPS is π/4-DQPSK. IS-95 in the US uses the same uplink and downlink frequency bands, but channels, codes in this case, occupy much wider 1.25 MHz frequency bands. It uses QPSK for downlink communications and OQPSK for uplink transmission. In Europe, the bands assigned to, and used by, GSM are 890–915 MHz in the uplink direction and 935–960 MHz in the downlink direction. The 25 MHz bands available in each direction are further broken into 124 frequency channels of 200 kHz each. Each frequency channel in turn is accessed by up to eight users, using time slot assignments. The modulation technique adopted is called 0.3 GMSK. Additional wireless communication channels for cellular communications have been made available in the US in the 1.85–1.99 GHz band, the so-called PCS band, and bands ranging from 1.71 to 1.9 GHz in Europe. For simplicity in providing examples in this chapter, we focus on the modulation techniques used in the 800–900 MHz bands. The discussion in this chapter will focus on the digital modulation techniques namely DQPSK, GMSK, QPSK, and OQPSK will be described in some detail. It is to be noted, however, that multiple criteria are involved in the specific choice of a modulation technique to be used. Examples of such criteria include bandwidth efficiency, power efficiency, cost and complexity, and performance in fading channels. Note that the battery-operated mobile or subscriber terminal, in particular must be reasonably inexpensive, small in size, and parsimonious in its use of power. Operation in a fading environment means that constant amplitude modulation techniques with nonlinear amplifiers used are favored. This is one reason for the original choice of FM for the first-generation analog systems. Basic digital modulation techniques that include on–off keying (OOK) or amplitude-shift keying (ASK), phase-shift keying (PSK), and frequency-shift keying (FSK) are first described. We will then touch briefly on quadrature amplitude modulation (QAM), used, incidentally, in wireline modems to increase the transmission bit rate over a bandwidth-limited channel, such as the telephone access line from home or office to a telephone exchange or central office. Quadrature phase-shift keying (QPSK) is a special case of QAM. We will go on to discuss enhanced digital modulation techniques such as OQPSK or MSK, and GMSK, chosen to work relatively well in the constrained bandwidth, fading environment of wireless cellular systems.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

186

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

INTRODUCTION TO DIGITAL MODULATION TECHNIQUES Consider an unmodulated sinusoidal carrier A cos  0 t operating continuously at a carrier frequency of f0 Hz, or  0  2  f 0 radians/sec. Its amplitude, frequency, and phase may

Figure A.1 OOK signal corresponding to binary 1, 0, 1. individually be varied or modulated in accordance with an information-bearing signal to be transmitted to provide, in the case of the first two, the well-known AM and FM high frequency signals used in (analog) radio broadcast systems with which we are all familiar. The simplest digital equivalents are called, respectively, as noted above, on-off keyed or amplitude-shift keyed transmission, OOK or ASK, frequency-shift keyed transmission, FSK, and phase-shift keyed transmission, PSK. In this digital case, successive binary digits (bits), 0 or 1, are used to vary or modulate the amplitude, frequency, and phase, respectively, of the unmodulated carrier. The binary, informationbearing, digits are also referred to as comprising the baseband signaling sequence. In the simplest case, that of OOK transmission, a binary 0 will turn the carrier off; A cos  0 t will be transmitted whenever a binary 1 appears in the baseband signaling sequence. Say the binary sequence is being transmitted at a bit rate of R bits/sec. Then each binary symbol or bit lasts 1/R sec. An example of an OOK signal corresponding to the three successive bits 1, 0, 1 appears in Fig. A.1. Note that OOK transmission is not suitable for digital mobile systems because of the variation in amplitude of the signals transmitted. Note that constant-amplitude signals are preferred because of the fading environment, which, by definition, introduces unwanted random amplitude variations. In actuality, the instantaneous change in amplitude shown occurring in Fig. A.1, as the OOK signal turns off and then comes back on again, cannot occur in the real world. Such abrupt transitions would require infinite bandwidth to be reproduced. In practice, the signals actually transmitted are shaped or prefiltered to provide a more gradual transition between a 0 and a 1, or vice-versa. The resultant transmitted signal corresponding to a baseband binary 1 may be written A h ( t ) cos  0 t . The low-frequency time function h(t) represents a signal-shaping function. This signal shaping is designed to allow the transmitted OOK sequence to fit into a prescribed system bandwidth with little or no distortion. The rf carrier transmission bandwidth, the frequency spread of the modulated transmitted signal about the carrier, is readily shown to be given by 2B, with B the baseband bandwidth, the bandwidth required to transmit the baseband binary sequence of 1s and 0s. Note that the baseband bandwidth, in Hz, varies between R/2 and R for the type of shaping discussed there. Hence the transmitted signal bandwidth varies between R and 2R Hz. As an example, if a 10 kbps digital information signal is to be transmitted over the wireless radio channel, the baseband bandwidth ranges from 5 kHz to 10 kHz, depending on the shaping used. The radio channel bandwidth would range from 10 to 20 kHz. Since OOK transmission, even with signal shaping, results in amplitude variation and is hence undesirable for mobile communication, we look to other types of digital modulation to provide the desired digital signaling. Most implementations in practice use some variation of phase-shift keying or PSK. However, we continue our discussion by first considering frequency-shift keying or FSK, the binary version of FM. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

187

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

In the case of FSK, the carrier frequency shifts between two different frequencies f 1  f 0   f and f 2  f 0   f , depending on whether a 0 or a 1 is being transmitted. The

term  f is called the frequency deviation about the average carrier frequency f0. An example appears in Fig. A.2. It is then found that the transmitted signal bandwidth is given approximately by 2  f  2 B , where the baseband bandwidth B again varies between R/2 and R, depending on the shaping used. The FSK transmission bandwidth can thus be substantially more than the bandwidth required for OOK transmission. The same result is obtained for analog

Figure A.2 FSK signal. FM. A simple pictorial example of the FSK spectrum, indicating the carrier frequencies and the transmission bandwidth, appears in Fig. A.3. Consider now phase-shift keying, or PSK, the last of the three basic ways of digitally modulating a carrier in conformance with the information-bearing binary signal sequence. In this case, as an example, we might have the carrier term A cos  0 t transmitted when a binary 1 appears in the baseband binary sequence; - A cos  0 t = A cos(  0 t   ) when a 0 appears. An example appears in Fig. A.4. PSK transmission turns out to be the best scheme to use in the presence of noise. It requires a phase reference at the receiver, however, in order for accurate, error-free, detection of the transmitted binary signal sequence to be carried out. Note that both FSK and PSK provide ostensibly constant-amplitude transmission, a desirable property in the presence of signal fading with its attendant random variation of amplitude. However, note from Fig. A.4 that abrupt phase changes occur when the signal switches from a 1 to a 0 and vice-versa.

Figure A.3 FSK spectrum.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

188

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure A.4 PSK signal. Note that, the phase changes do result in an undesirable increase in signal transmission bandwidth and a resultant undesirable variation in the transmitted signal amplitude, because of the limited bandwidth encountered during transmission. Steps thus have to be taken to reduce these potentially large, abrupt phase changes. SIGNAL SHAPING Recall that shaping must be used in digital communication schemes to keep the bandwidth of the resultant modulated transmission signal within the prescribed system bandwidth. The discussion in this section should provide the necessary insight to understand signal shaping in general. The shaping of a high-frequency radio carrier signal at frequency f0 is indicated by writing the carrier signal in the form h(t) cos  0 t . The time-varying function h(t) is the shaping function. To keep the modulated signal smooth, particularly during bit transitions from a 1 to a 0 and vice-versa, it is clear that the function h(t) must have a form such as that shown in Fig. A.5. It should be maximum at the center of the binary interval 1/R and then drop off smoothly on either side of the maximum. One can then reproduce the original binary symbol, whether a 1 or a 0, by sampling the value at the center. Note, as an example, that if such a function were to multiply the OOK sinewave, it would reduce considerably the abrupt amplitude changes encountered in

Figure A.5 Typical shaping function. switching from a 1 transmitted to a 0 and vice-versa. Rather than focus on the modulated highfrequency signals, and the effect of multiplying them by h(t), however, it is simpler to consider shaping of the binary baseband signal sequence of 1s and 0s. The multiplication by cos  0 t then serves to shift the shaped baseband spectrum up to a spectrum centered at the carrier frequency f0. Shaping can thus be carried out either at baseband, for conceptual simplicity, or directly at the carrier frequency. The result is the same. Thus, consider a baseband sequence of 1s and 0s, corresponding to a bit rate of R bits/sec (bps), each bit multiplied by h(t), as shown in Fig. A.6. The 1s are each represented by the appearance of h(t). The 0s give rise to a gap in the time sequence. It is well known from Fourier analysis that as a time function is narrowed, its spectral bandwidth increases correspondingly. As the function broadens in time its bandwidth decreases. The width of a pulse and its bandwidth are inverse to one another. Thus if the width of h(t) in Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

189

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Fig. A.6 is reduced, the bandwidth of the baseband signal sequence is increased; if the width of h(t) increases, the corresponding bandwidth decreases. As the width of h(t) increases however, the pulses begin to intrude into the adjacent binary intervals. This gives rise to inter-symbol interference, as shown in Fig. A.6. Initially there is no problem. For it is always possible to sample a binary pulse at its center and determine whether it is a 1 or 0. But eventually the inter-symbol interference becomes large enough to affect the value of the binary signal at the center of its particular interval, and mistakes such as confusing a 1 for a 0 and vice versa may occur. This is particularly true if noise is present, interfering with the signal transmitted. There is thus a tradeoff between inter-symbol interference and the bandwidth needed to transmit a sequence of pulses. The same observation applies to the modulated binary sequence obtained by multiplying by cos  0 t and shifting up to the carrier frequency f0.

Figure A.6 Binary sequence.

190

Consider now the more general case of sinusoidal rolloff shaping. This is used to control the inter-symbol interference, while keeping the bandwidth required for transmission as low as possible. By sinusoidal

Figure A.10 Sinusoidal rolloff shaping spectrum. roll off shaping we mean shaping with the spectrum H   given by

The corresponding pulse shape h(t), the inverse Fourier Transform of H(ω), may be shown to be given by Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The radian bandwidth is given by 2πB = ωc + ωx. In units of Hz we get B = fc + fx, with the two frequency parameters fc and fx obtained by dividing the corresponding radian frequency terms by 2π. The sin ωct/ωct term indicates that successive binary digits may still be spaced 1/R = π/ωc sec apart with 0 inter-symbol interference. The bandwidth B = fc + fx required to transmit these pulses is commonly written in a form involving the ratio fx/fc, defined to be the rolloff factor r. Note from Fig. A.10 that this ratio determines the rate at which the shaping spectrum drops from its maximum value of 1 to 0. As fx, or, equivalently, r, decreases, approaching 0, the spectrum drops more rapidly, reducing the bandwidth B, but increasing the possibility of inter-symbol interference due to timing jitter. Note that the bit rate R, given by R=ωc/π =2fc, since ωc =2πfc. This enables us to write the bandwidth B in the form

Here the rolloff factor r ≡ fx/fc ≤ 1. The transmission or radio bandwidth for OOK or PSK transmission is then 2B = R (1+ r). The transmission bandwidth in these cases thus ranges from values just above R to 2R. Labeling the transmission bandwidth BT, we have 191

As an example, that 14.4 kbps transmission is desired. If sinusoidal rolloff shaping and PSK modulation are used, the required transmission bandwidth is 21.6 kHz if a rolloff factor of 0.5 is used; 28.8 kHz if the rolloff factor is 1. If 28.8 kbps is desired, these bandwidths double. Conversely, given the bandwidth available, one can determine the maximum binary transmission rate for a given rolloff factor. The second-generation cellular wireless system D-AMPS (IS-136), as an example, uses sinusoidal rolloff shaping, with a rolloff factor r = 0.3A. It slots its overall frequency band assignment into 30 kHz-wide user bands, as noted earlier. If PSK modulation were to be used, the maximum binary transmission rate would be R = 30,000/1.35 = 22.2 kbps. The modulation scheme actually used is labeled DQPSK. The actual transmission rate is 48.6 kbps. The Japanese PDC system also uses sinusoidal rolloff shaping, with the rolloff factor r = 0.A. The bandwidth available per user is 25 kHz. For PSK modulation, the binary transmission rate would then be 16.7 kbps. This system also uses DQPSK and the actual binary transmission rate is 42 kbps. MODULATION IN CELLULAR WIRELESS SYSTEMS Recall on digital modulation schemes that specific problems arise in dealing with transmission over a bandwidth-limited cellular wireless channel, which has the potential for introducing random amplitude variations due to fading. Modulation schemes must be selected that provide an essentially constant amplitude of transmission. PSK and FSK do have this property, but instantaneous phase shifts of up to π radians (180◦) that may be incurred in changing from one bit value to another give rise to amplitude changes during transmission due to limited bandwidth. In addition, the bit rate possible using PSK over a band limited channel of bandwidth B is strictly limited by B and directly proportional to it. We thus look to somewhat more complex digital modulation schemes that produce transmitted signals with limited amplitude and phase variations, from one bit to another, while allowing higher bit rates to be used for a given bandwidth. These types of modulation are based on, and utilize, phase-quadrature

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

transmission. We thus begin the discussion in this section by focusing on a technique called quadrature PSK or QPSK, which allows a system to transmit at double the bit rate for a given bandwidth. Quadrature PSK (QPSK) Consider transmitting two carriers of the same frequency, f0, at the same time, with one in phase quadrature to the other: cos  0 t and sin  0 t . Let each have two possible amplitude values, say  1 , for simplicity. Let each last the same length of time, T sec. Each then corresponds to a PSK signal. Let a i   1 be the value assigned to the cosine (inphase) carrier of the ith such signal; let b i   1 be the value assigned to the sine (quadrature) carrier of the same signal. Say each of the two carriers is shaped by a shaping factor h(t). The ith signal si(t), lasting an interval T, is then given by the expression

The shaping factor h(t) is assumed here to be centered about the middle of the interval T. This expression may also, by trigonometry, be written in the following amplitude/phase form

The four possible values of the amplitude-phase pair (ri, θi) are obviously related one-to-one to the corresponding values of (ai, bi). Four possible signals are thus obtained, depending on the values of the combination (ai, bi), each one of which requires the same bandwidth for transmission. How are the actual values of ai and bi determined? How would transmission be carried out? Let the bit rate be R bps, as previously. Successive pairs of bits are stored every T sec, giving rise to four different two-bit sequences. Note then that we must have the two-bit interval T = 2/R. Each of these four sequences is mapped into one of the four carrier signals, and that particular signal, lasting T sec, is transmitted. One possible mapping of these signals appears in Table A.1. A block diagram of the resultant QPSK modulator appears in Fig. A.11. If we use sinusoidal roll-off shaping, the baseband bandwidth, B = (1/T) (1 + r), with 1/T replacing the previous R/2. The bandwidth required to transmit a given bit sequence has thus been reduced by a factor of two. Conversely, given a bandwidth B, we can now transmit at twice the bit rate. The transmission or rf bandwidth, as in the case Table A.1 Example of mapping, binary input sequence →QPSK (ai, bi).

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

192

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure A.11 Generation of QPSK signals. of PSK, is double the baseband bandwidth, or (2/T)(1 + r). Demodulation, the process of reconstructing the original binary sequence at a receiver, is readily carried out by noting that the cosine and sine terms in the QPSK signal that are orthogonal to one another. Multiplying the QPSK received signal in two parallel operations by cosine and sine terms, respectively, each exactly in phase with the transmitted cosine and sine terms, and then integrating over the T-sec interval (effectively low-pass filtering the two multiplied terms), will extract each term separately. This operation requires exact phase knowledge of the transmitted signal, precisely the same problem encountered with PSK transmission. An alternative procedure is to use successive signals as phase references for the signals following. The operation of quadrature modulation may be diagrammed geometrically on a two dimensional diagram, as shown in Fig. A.12 (a). The horizontal axis represents the coefficient ai of the inphase, cosine; the vertical axis represents bi, the coefficient of the quadrature, sine term. Note that putting in the four possible values of (ai, bi), Table A.2 π/4 DQPSK.

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

193

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure A.12 QPSK signal constellation one obtains the four points of operation shown in Fig. A.12(a). These points are referred to as the signal constellation of the QPSK signals. Each point may be written as the complex number or phasor ai, the real part of which is ai, the imaginary part of which is bi. A little thought will make it clear that an equivalent representation of the QPSK signals would be obtained by rotating the signal constellation of Fig. A.12(a) 45◦ or π/4 radians, producing the constellation shown in Fig. A.12(b). This constellation corresponds to transmitting one of the four signals  cos ω0t,  sin ω0t, depending on the particular value of ai = (ai, bi). It thus represents another mapping of the four possible values of (ai, bi), the first example of which was provided in Table A.2. Note that, although the four points in the signal constellation lie on a circle, transitions between these points, and hence signals transmitted, will result in some signal amplitude variation in the vicinity of the transition times. The T-sec long interval we have introduced, corresponding to the time to store two bits, and the time over which each QPSK signal (one of four possible) is transmitted, is called the baud or symbol interval, as distinct from the binary interval 1/R. One refers to a given number of baud, rather than bits/sec, and it is the baud number that determines the bandwidth. In the special case of two-state, binary, transmission the number of baud and the bit rate are the same. For the QPSK case, however, they differ by a factor of 2. 8-PSK Consider now extending this idea of storing two bits to reduce the bandwidth required for transmission. Say three successive bits are stored, as an example. This results in 23 = 8 possible 3-bit sequences. Corresponding to these sequences, we could transmit eight different signals, all of the same amplitude but differing in phase by π/4 radians. Alternately, these eight signals could differ in both phase and amplitude. The former case is referred to as 8-PSK signaling. This type of signaling would be represented by a signal constellation similar to that of Fig. A.12 with eight points symmetrically located on a circle centered around the origin. This 8-PSK signaling scheme has, infact, been adopted for an enhanced bit-rate version of GSM called EDGE. This The second case noted above, when designing a set of eight signals, each corresponding to one of the 3-bit input sequences, is to have both amplitude and phase differ. The only difference with the QPSK and the 8-PSK cases is that a signal constellation consisting of a mapping of eight possible values of the amplitude pair (ai, bi) is obtained, each of which may be defined as a complex number with ai its real part in the plane of the signal constellation and bi its imaginary part. Quadrature Amplitude Modulation (QAM) More generally now, say k successive bits are stored, resulting in M = 2k possible k-bit sequences. The resultant set of M different signals must then differ in the values of ai and bi. This modulation technique is referred to as quadrature amplitude modulation, or QAM. These signals correspond to M possible points in a two-dimensional signal constellation. QPSK and 8-PSK are clearly special cases, with the signal points arranged on a circle. A specific example of a 16-point signal constellation appears in Fig. A.13. In this case, one of 16 possible signals would be transmitted every T-sec interval. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

194

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Four successive bits would be stored in this case to generate the corresponding 16-QAM signal. The bandwidth required to transmit this sequence of signals is one-fourth that of the equivalent PSK signals, since the length of a given one of the 16 possible signals has been increased by a factor of four. Conversely, given a specified channel bandwidth one can transmit at four times the bit rate. More generally, with k successive bits stored, the

Figure A.13 Signal constellation 16-QAM. baud length T increases by a factor of k, allowing k times the bit rate to be transmitted over a given bandwidth. Aside from the special case of 8-PSK, QAM has not generally been adopted for cellular wireless systems, because it does introduce amplitude variations in the resultant signal set transmitted. It has been adopted for wireless local area networks (LANs. In particular, QAM is the basis of operation of the 28.8 kbps telephone-wire modems (from modulator–demodulator) that were in use prior to the introduction of the 56 kbps modems commonly used at this time. Minimum-Shift Keying (MSK) This modulation scheme doubles the bit rate possible over the band-limited wireless channel, a desirable feature in this environment. There is still the problem of an abrupt 180◦ (π rad.) phase shift occurring during the switch from one QPSK signal to another at intervals of T sec, a problem we noted existed with PSK, if both ai and bi change sign. This is apparent from an examination of the two constellations of Fig. A.12. Note that the phase changes by 90◦ (π/2 rad.) if only one of these coefficients changes. A scheme developed to reduce the maximum phase shift, and hence keep the amplitude quasi-constant during transmission over the band-limited wireless channel, is called offset QPSK (OQPSK) or, in a modified form, minimum-shift keying (MSK). This scheme serves as the basis for modulation schemes selected for some of the current cellular wireless systems. In OQPSK the modulation by one of the two (quadrature) carriers is offset in time by one bit. (T/2 = 1/R sec.) A block diagram appears in Fig. A.15. MSK is similar, except that shaping or multiplying by the sine function cosπt/T is first carried out, before carrying out the multiplying (modulation) and delay functions shown in Fig. A.15. This is shown schematically in the block diagram of Fig. A.16. Note that the acronym “lpf ” shown there represents the shaping by cos πt/T. It stands for low-pass filtering. The system to the left of

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

195

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure A.14 Telephone modem bandwidth

196

Figure A.15 OQPSK block diagram; T = baud or symbol interval

Figure A.16 MSK modulation. the two input lines is again the same as that appearing in Fig. A.11(a). The (ai, bi) pair again represents the mapping of two successive information bits into one of four possible pairs, an example of which appeared in Table A.1. The MSK waveform may therefore be written in the form

The MSK signal is thus similar to the QPSK signal, except that the sin πt/T shaping function multiplying the quadrature carrier term is in phase quadrature to the cos πt/T shaping function used to multiply the inphase carrier term. The sin πt/T term arises simply because of the T/2 = 1/R delay

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

following the lpf shaping in the quadrature section of the modulator. Note that the coefficients ai and bi change every T = 2/R sec. Mathematically,

Note from this form of the MSK signal si(t) that the signal has the desired constant amplitude. Because of the T/2 delay separating the influence of ai and bi (Fig. A.16), only one of these coefficients can change at a time. The maximum phase change possible is thus π/2 radians, rather than the maximum phase change of π radians in the QPSK case. But why the minimum shift keying designation? This is explained by rewriting the expression for the MSK signal in yet another form

This form is representative of frequency-shift keying (FSK), with frequency deviation  f , in Hz, and frequency spacing, in Hz, given by 2  f = 1/T = R/2, R the information bit rate. It is in this context that the minimum shift keying appellation arises. Gaussian MSK (GMSK) Now consider gaussian MSK, or GMSK, used in the GSM mobile wireless system. This modulation scheme extends MSK to include gaussian shaping. Figure A.18 demonstrates the process. In Fig. A.18(a), we show the FSK equivalent of MSK. In GMSK, a low-pass gaussian shaping filter h(t) precedes the FSK modulator, as shown

Figure A.18 Gaussian MSK modulation

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

197

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

Figure A.19 Gaussian shaping. in Fig. A.18(b). This shaping filter has the specific gaussian form below, hence its name.

This shaping is to be applied to each information bit input, as shown in Fig. A.18(b). Note that the parameter σ controls the width of the resultant shaped binary pulse entering the modulator. Increasing σ increases the spread of the pulse into adjacent bit intervals, increasing the inter-symbol interference. Unlike the case of sinusoidal rolloff shaping, where the shaping function is specifically chosen to attain zero intersymbol interference, except for the effect of timing jitter, here controlled inter-symbol interference is introduced. The reason is that by increasing the shaped binary pulse width, the transmission bandwidth is thereby reduced: this represents the well-known inverse relationship between pulse width and bandwidth. This inverse relationship is demonstrated in this specific case by calculating the Fourier Transform of h(t), its spectrum. It is readily shown that this transform, H(ω), is also gaussian shaped and is given by following C is a constant. The 3 dB bandwidth, i.e., the bandwidth in Hz at which H(ω) is 1/√2=0.707 of its peak value, is readily shown to be given by B = 0.133/σ . Note the inverse relation between pulse width σ and bandwidth B. This shaping function h(t) and its transform H(ω) are sketched in Fig. A.19. An alternate representation of the shaping function h(t) may be written, replacing σ by its equivalent in terms of the bandwidth B

Note that as σ→0or B→∞, the shaping effectively disappears and we recover the original MSK signal at the modulator output. The effect of the shaping function on the input to the FSK modulator of Fig. A.18(b) is found by considering an unshaped square-topped binary pulse, representing a binary 1, say, passing through the (low-pass) shaper before being acted on by the modulator. This pulse, of width 1/R = T/2 ≡ D, excites the gaussian filter of Fig. A.18(b). The desired output g(t), which is, in turn, the input to the FSK (MSK) modulator, is found by convolving the two time functions. The resultant time function is found to be given by (A.13), suppressing a multiplicative constant

The complementary error function, erfc(x), appearing in (A.13) is well-known as the integral of the gaussian function

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

198

R&D NEXTEVOLUTION

A Contribution of R&D NEXTEVOLUTION

Branding Intellects

The effect of the gaussian shaping may be determined by plotting the shaped time function g(t) as a function of the normalized bandwidth BN ≡ BD= BT/2 = B/R (Steele, 1992: 535). As BN decreases, reducing the bandwidth with the binary input rate R fixed, g(t) spreads more into adjacent binary intervals, increasing the inter-symbol interference. The resultant spectrum, however, the Fourier Transform of g(t), or, equivalently, the effect of multiplying H(ω) by the sin x/x spectrum of the square binary input pulse, turns out to have reduced spectral lodes as compared with the MSK spectrum. GMSK thus

effectively provides a reduced transmission bandwidth as compared with plain MSK, at the cost of intersymbol interference. A good compromise value of BN turns out to be BN = 0.3. This value provides a substantial reduction in transmission bandwidth with not too much inter-symbol interference introduced. This is the value that has, in fact, been adopted for the GMSK modulation used in GSM systems. π/4-DQPSK (differential QPSK) The designers of IS-136 (previously known as IS-54) or D-AMPS, the second generation North American TDMA cellular system, chose for its modulation scheme π/4-DQPSK (differential QPSK). This scheme uses eight different phase positions, with constant amplitude, as its signal constellation. The four different values of the (ai, bi) pair discussed previously are assigned to corresponding differential changes in the constellation phase positions. Given a current point in the constellation at which the system is operating, say at the end of the T-sec interval n, there are four different points to which the system can move, depending on which of the four values of (ai, bi) apears in the next T-sec interval, n + 1. Table A.2 provides the necessary mapping of (ai, bi) to the differential phase changes, θ(n) = θ(n + 1) – θ(n). A diagram of the eight point constellation, with an example of the differential phase changes indicated, appears in Fig. A.20. In this example, the system phase point at the end of interval n is taken to be 3π/4, as shown in the figure. The pair (ai, bi) = (−1,−1) arising in the next T-sec interval would then change the phase by −3π/4 rad., rotating the phase point to θ(n + 1) = 0, as shown in the figure. The three other possible rotations, corresponding to the three shown in Table A.2, are indicated in Fig. A.20. The transmitted signal is then of the constant-amplitude phase shift type using one of the eight possible phase positions. Transitions between signal points will result in amplitude variations in the successive signals transmitted, just as in the case

Figure A.20 π/4 DQPSK. of QPSK. These variations, just as those for 8-PSK, will be reduced as compared with QPSK, however. How would detection at the receiver be accomplished in this case? It is clear that phase detection is necessary, since the information to be reproduced is carried in the phase of the received signal. Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013

199

R&D NEXTEVOLUTION

Branding Intellects

A Contribution of R&D NEXTEVOLUTION

The receiver must generally be locked in precisely to the phase and frequency of the transmitted signal in the case of phase-shift keyed and FSK-type modulation. Such techniques then require the receiver to correctly track the absolute phase of the transmitted signal for correct detection to take place. Such absolute phase information is not needed, however, in using this type of differential phase modulation scheme. Differential phase shift decisions only need be made, each T-sec interval, to correctly determine the (ai, bi) pair generated at the transmitter, from which, in turn, the corresponding binary information sequence may be reconstructed. In essence, the transmitted signals, one after the other, provide the necessary phase reference for the signal following. This can simplify the phase detection process considerably. Provision must be made, however, for detecting and correcting an erroneous phase decision; otherwise the phase error could perpetuate indefinitely.

200

Cellular Mobile Communication- A Fundamental Perspective Lecture note is prepared by Rony Kumer Saha for class room use only

edition 02, version 02 2013