sensors Article
Low Power Near Field Communication Methods for RFID Applications of SIM Cards Yicheng Chen 1 , Zhaoxia Zheng 2 , Mingyang Gong 2 and Fengqi Yu 1, * 1 2
*
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China;
[email protected] School of Optical and Electronic Information, Huazhong University of Science & Technology, Wuhan 430074, China;
[email protected] (Z.Z.);
[email protected] (M.G.) Correspondence:
[email protected]; Tel.: +86-130-2544-7469
Academic Editor: Vittorio M. N. Passaro Received: 25 February 2017; Accepted: 13 April 2017; Published: 14 April 2017
Abstract: Power consumption and communication distance have become crucial challenges for SIM card RFID (radio frequency identification) applications. The combination of long distance 2.45 GHz radio frequency (RF) technology and low power 2 kHz near distance communication is a workable scheme. In this paper, an ultra-low frequency 2 kHz near field communication (NFC) method suitable for SIM cards is proposed and verified in silicon. The low frequency transmission model based on electromagnetic induction is discussed. Different transmission modes are introduced and compared, which show that the baseband transmit mode has a better performance. The low-pass filter circuit and programmable gain amplifiers are applied for noise reduction and signal amplitude amplification. Digital-to-analog converters and comparators are used to judge the card approach and departure. A novel differential Manchester decoder is proposed to deal with the internal clock drift in range-controlled communication applications. The chip has been fully implemented in 0.18 µm complementary metal-oxide-semiconductor (CMOS) technology, with a 330 µA work current and a 45 µA idle current. The low frequency chip can be integrated into a radio frequency SIM card for near field RFID applications. Keywords: subscriber identity module (SIM); near field communication; radio frequency identification; low-power design
1. Introduction As communication and computing performance evolves, smart phones with various types of sensors have the potential to become identification cards and credit cards. Radio frequency identification (RFID) [1], micro payment [2,3], physical access control [4], location and activity recognition applications [5,6] based on mobile phones are proposed. This is so that the users have no need to keep more cards and keys in their wallet or pocket as the number of services increases. The 13.56 MHz near field communication (NFC) technology based on ISO 14443 protocol is the popular scheme for mobile phones, and this uses the NFC chip, subscriber identity module (SIM) cards, and the single wire protocol (SWP) interface [3,7]. Secure element (SE) chips in the SIM card are used for security assurance, and have usually passed the Common Criteria evaluation assurance level (EAL) 5+ standard and have the same security level as bank smartcards [8]. The 13.56 MHz NFC technology in mobile phones is confronted by two problems in many usage scenarios. Firstly, it is not suitable for middle and long distance RFID applications, such as garage door control and indoor location tracking [9–11]. Secondly, the metallic shells of phones can cause a frequency shift for 13.56 MHz technology and a reduction of the field intensity, which can affect the communication quality for NFC devices [12,13]. To solve these problems, 2.45 GHz radio frequency (RF) Sensors 2017, 17, 867; doi:10.3390/s17040867
www.mdpi.com/journal/sensors
Sensors 2017, 17, 867
2 of 14
chips are introduced into the mobile phones; in particular, secure digital (SD) cards and SIM cards are used as substrate for 2.45 GHz chips and antennas [14–16]. When a 2.45 GHz RF chip is integrated into the SIM card, there are still two crucial problems for RFID SIM cards: high power consumption and an uncontrollable working distance of up to tens of meters for the 2.45 GHz chip. How to deal with the dilemma between the short distance and long distance applications is a difficult challenge. This paper presents a strategy to tackle the dilemma, and an improved constitution of RFID SIM cards is proposed. In this type of RFID SIM card, a 2.45 GHz RF chip and an SE chip are integrated. However, the 13.56 MHz technology is not chosen for RFID SIM cards, because the coil of 13.56 MHz is fairly large and is difficult to integrate into SIM cards, especially nano-SIMs. In addition, the metallic shells and card slots of mobile phones can affect the communication quality of 13.56 MHz. Therefore, a low power near-distance communication chip is discussed and customized. Compared to 13.56 MHz NFC chips, the customized low frequency (LF) chip can get energy directly from the mobile phone, so there is no need for the complex amplitude limiter, rectifier, and voltage stabilizing circuits. Besides this, the LF chip is implemented with the function of receiving data only, which results in the removal of transmitting circuits and the reduction of power consumption. The remainder of this article is organized as follows. Section 2 compares the ordinary SIM and RFID SIM, and describes an improved RFID SIM structure including the low frequency chip. Section 3 briefs the basics of the low frequency data transmission model. The real low frequency communication system design is discussed in Section 4, which describes the frontend circuit structure of the LF chip, the implementation of programmable gain amplifiers, and an improved differential Manchester decoder. The low frequency chip is implemented and tested; the results are shown in Section 5. Finally, concluding remarks are made in Section 6. 2. An Improved RFID SIM Structure Traditional SIM, machine-to-machine (M2M) SIM [17,18], and e-SIM cards [19–22] perform user identification, M2M communication, and remote provisioning by the SE chip. The SE chip incorporates a secure ARM processor, embedded flash, cryptographic engines, and some interfaces. One of the interfaces is compatible with the ISO/IEC 7816-3 standard for the communication between the mobile phone and the SIM card. The first generation RFID SIM integrated a secure chip, a 2.45 GHz RF chip, and a 2.45 GHz RF antenna [14,16]. The SE chip configures and controls the 2.45 GHz RF chip via a serial peripheral interface (SPI). It works well in many middle and long distance RFID applications, such as staff attendance. However, for some short distance scenarios, for example, ticket payments for buses and subways, the transaction distance of the micro-payment system must be restricted to about 10 cm. It is a crucial challenge for 2.45 GHz RF communication to make sure that the transactions do not occur beyond a certain range. One scheme is that a near field state indicator can be applied to inform the SE chip and the RF chip. As such, second generation RFID SIM cards and dual-mode readers have been designed, as shown in Figure 1. The dual-mode reader consists of a high frequency reader and a low frequency reader, which can transmit 2.45 GHz and 2 kHz electromagnetic signals. The 2.45 GHz RF chip, the SE chip, the low frequency chip, and the 2.45 GHz and LF antennas constitute the RFID SIM. This type of RFID SIM can be embedded in common SIM, M2M, and e-SIM cards. The SE chip deals with the subscriber identity operations and controls a 2.45 GHz and low frequency chip. The LF chip can only receive data within a short communication distance and with less working current. The 2.45 GHz RF chip can transmit and receive data through different frequency channels and signal strengths.
Sensors 2017, 17, 867
3 of 14
Sensors 2017, 17, 867
3 of 14
Sensors 2017, 17, 867
3 of 14
Figure 1. The dual-mode reader and radio frequency identification (RFID) SIM card. Figure 1. The dual-mode reader and radio frequency (RFID) card.periodically and In middle and RFID applications, the 2.45identification GHz RF chip is SIM woken Figure 1.long The distance dual-mode reader and radio frequency identification (RFID) SIM card. is sent identification data during its working states. For short distance applications, the typical In middle and long distance RFID applications, the 2.45 GHz RF chip is woken periodically and process is as follows: In and long distance RFIDitsapplications, the For 2.45short GHz distance RF chip is woken periodically is middle sent identification data during working states. applications, the typicaland is processthe is as follows: sent data during its working states. For short typical process is (1) identification First, 2.45 GHz RF chip sleeps with about 1 µAdistance current,applications, and the LF the reader sends 2 kHz as follows: signals; (1) First, the 2.45 GHz RF chip sleeps with about 1 µA current, and the LF reader sends 2 kHz (2) Secondly, the LF chip determines whether or not a card approaches, receives the LF data, and (1) First,signals; the 2.45 GHz RF chip sleeps with about 1 µA current, and the LF reader sends 2 kHz signals; wakes up the the 2.45LF GHz (2) Secondly, chipchip; determines whether or not a card approaches, receives the LF data, and (2) Secondly, the LF chip determines whether or not a card approaches, receives the LF data, wakes the GHz 2.45 GHz (3) Finally, theup2.45 chipchip; requests the reader to suspend the transmission of LF signals and and Finally, wakes up the 2.45 GHz chip; (3) the 2.45 GHz chip requests the reader to suspend the transmission of LF signals and handles 2.45 GHz transactions. (3) Finally, the 2.45 2.45GHz GHztransactions. chip requests the reader to suspend the transmission of LF signals and handles handles 2.45 GHz 3. Low Frequency Datatransactions. Transmission Model
3. Low Frequency Data Transmission Model
3. Low Frequency Datathe Transmission Model process between the reader and the RFID SIM card. In Figure 2 describes data transmission Figure 2 describes the data transmission process between the reader and the RFID SIM card. In the low frequency mode, the raw data is and modulatedbyby microcontroller (MCU) the low2frequency the transmission raw data isencoded encoded and modulated thethe microcontroller unit unit (MCU) Figure describesmode, the data process between the reader and the RFID SIM card. In the in theinreader. TheThe waveforms generated by MCU areamplified amplified operational amplifiers, the reader. waveforms generated by the the MCU are byby operational amplifiers, and and low frequency mode, the raw data is encoded and modulated by the microcontroller unit (MCU) in then then transmitted through thethe antenna. The electromotive forceininthethe coils of the RFID SIM card is transmitted through antenna. The electromotive force coils of the RFID SIM card is then the reader. The waveforms generated by the MCU are amplified by operational amplifiers, and generated by electromagnetic induction. operationfrequency frequency is distinctly lower and the generated by electromagnetic induction.Because Because the the operation is distinctly lower and the transmitted through the antenna. The electromotive force in the coils of the RFID SIM card is generated in the cards muchsmaller smaller compared compared to at 13.56 MHz, the the coils coils in the SIMSIM cards arearemuch to the theNFC NFCtechnology technology at 13.56 MHz, by electromagnetic induction. Because the operation frequency is distinctly lower and the coils in the electromotive signal is very weak andneeds needsto to be be handled handled with Suitable matching and filter electromotive signal is very weak and withcaution. caution. Suitable matching and filter SIM cards arebetween much smaller compared to the NFC technology at 13.56 MHz, the electromotive circuits the coils and the LF chip should be designed carefully, as these become ansignal circuits between the coils and the LF chip should be designed carefully, as these become an is very weak and needs to be handled with caution. Suitable matching and filter circuits between influential factor on the working distance and decoding success ratio. After that, the voltage signal the influential factor on the working distance and decoding success ratio. After that, the voltage signal coils and the LF chip should befrom designed as thesetobecome influential factor on the working magnification is required, a few carefully, tens of microvolt severalan volts. The amplified signals are magnification is required, from a few tens of microvolt to several volts. The amplified signals are digitalized by a digital-to-analog and decoded through a differential Manchester distance and decoding success ratio. converter After that,(DAC) the voltage signal magnification is required, from a few digitalized by a digital-to-analog converter (DAC) and decoded through a differential Manchester decoder. tens of microvolt to several volts. The amplified signals are digitalized by a digital-to-analog converter decoder. (DAC) and decoded through a differential Manchester decoder.
Figure 2. The low frequency data transmission process.
Figure 2. The low frequency data transmission For simplicity, considering that low the reader antenna circular, theprocess. low frequency magnetic field Figure 2. The frequency data is transmission process. intensity (H) created by the reader is given as [23]
For simplicity, considering antenna is circular, the low frequency magnetic field considering that the reader = (1) + intensity (H) created by the reader is given given as as [23] [23] 2 q number of turns of the reader coil, R is the coil where I denotes the current of the reader coil, N is the = 2 3 (1) 2 + along radius, and X means the distance from center 2of (1) H =the I NR ( Rcoil 2 the + X 2 ) the direction of the X axis.
where I denotes the current of the reader coil, N is the number of turns of the reader coil, R is the coil where I denotes the current of the reader coil, N is the number of turns of the reader coil, R is the coil radius, and X means the distance from the center of the coil along the direction of the X axis. radius, and X means the distance from the center of the coil along the direction of the X axis.
Sensors 2017, 17, 867
4 of 14
For SIM cards, the magnetic field intensity (H) through the LF coil can be assumed to be equal at every point due to the small area and large number of turns of the LF coil. Assuming the SIM LF coil with the area S is parallel to the reader coil, the magnetic flux (Φ) of the LF coil can be expressed as Sensors 2017, 17, 867
Φ=
w
4 of 14
B·dS ≈ µHS
For SIM cards, the magnetic field intensity (H) through the LF coil can be assumed to be equal at every point due to the small area and large number of turns of the LF coil. Assuming the SIM LF coil means flux todensity, µ coil, denotes magnetic permeability, and is the inner with themagnetic area S is parallel the reader the magnetic flux (Φ) of the LF coil can be ·expressed as
(2)
where B product. When the number of turns of the LF coil is M, the induced voltage (V) of the LF coil by magnetic = ∙ S≈ (2) coupling is defined as where B means magnetic flux density, μ denotes magnetic qpermeability, and · is the inner product. LF coil is M, the 2induced voltage (V) 3of the LF coil by magnetic When the number of turns of the V = M dΦ dt ≈ MNµSR 2 ( R2 + X 2 ) dI dt coupling is defined as
(3)
= the induced ≈ (3) Equation (3) demonstrates that voltage 2 + of the LF antenna in the SIM card is proportional to the differential of the reader coil current. Accordingly, if the induced voltage (VSIM ) Equation (3) demonstrates that the induced voltage of the LF antenna in the SIM card is needs to satisfy a differential Manchester code format, theAccordingly, current (Ireader in the reader to be ) proportional to the differential of the reader coil current. if the) induced voltage coil (VSIMneeds needs to satisfy a differential Manchester code format,codes. the current (Ireader reader coil needs to the integral pattern of the corresponding Manchester Each bit) in ofthe differential Manchester code theµs integral of the and corresponding Manchester codes. Each bit of differential Manchester consumesbe 500 in thepattern SIM card, the corresponding saw tooth wave of the reader current has the code consumes 500 µs in the SIM card, and the corresponding saw tooth wave of the reader current same timing constraint, which is illustrated in Figure 3. has the same timing constraint, which is illustrated in Figure 3.
Figure 3. The 2 kHz transmission method for SIM cards by magnetic coupling.
Figure 3. The 2 kHz transmission method for SIM cards by magnetic coupling. We constructed an experimental prototype by using MCU, operational amplifiers, coils, capacitors, and resistors according to the above low frequency transmission model. The reader We constructed an experimental prototype by using MCU, operational amplifiers, coils, capacitors, current (Ireader) is generated according to Equation (3) and Figure 3. Because magnetic permeability in and resistors the abovethelow frequency transmission model. The reader current (Ireader ) air is according very small, into this method, induced electromotive signal is very weak, usually a few tens of is generated according to Equation (3) coil andis Figure 3.60Because in air is very micro volts. For example, the reader 60 mm by mm with magnetic 88 turns, andpermeability the coil in the SIM cardmethod, is 12.3 mmthe by induced 8.8 mm with 18 turns. The communication 8 cm between reader small, in this electromotive signal is verydistance weak, isusually a few the tens of micro volts. and the SIM card. The voltage peak-to-peak value of the reader is 3.8 V and the impedance of the For example, the reader coil is 60 mm by 60 mm with 88 turns, and the coil in the SIM card is 12.3 mm reader coil is 56 ohm. The induced voltage of the SIM card is about 50 µV according to Equation (3). by 8.8 mm with 18 turns. The communication distance is 8 cm between the reader and the SIM card. The signal cannot obtain enough voltage strength from the electromagnetic induction reader, so The voltage peak-to-peak value of the reader is 3.8 the impedance of the reader coil is 56 ohm. the LF chip needs powerful amplifiers to deal withVit.and LF communication is significantly different from common near field communication such as ISO/IEC 14443 and 15693. Furthermore, the signal The induced voltage of the SIM card is about 50 µV according to Equation (3). strengthcannot is influenced by the area and number strength of turns of from the reader and the SIM card induction coil, the The signal obtain enough voltage thecoil electromagnetic reader, working distance, and the angle between the reader coil and the SIM card coil. When the induced so the LF signal chip needs powerful amplifiers to deal with it. LF communication is significantly different is magnified significantly and enters a state of saturation, the influence of the above factors from common field communication such as approach ISO/IECneeds 14443 and 15693. Furthermore, can benear eliminated within a certain range. This to design appropriate amplifiers, the signal which is important the area working distance and the strength is influenced byforthe and number of decoding turns ofstability. the reader coil and the SIM card coil,
the working distance, and the angle between the reader coil and the SIM card coil. When the induced 4. Low Frequency Communication Implementation signal is magnified significantly and enters a state of saturation, the influence of the above factors can 4.1. The Frontend Circuit Design be eliminated within a certain range. This approach needs to design appropriate amplifiers, which is important forAccording the working and the to thedistance low frequency data decoding transmissionstability. model, the current (Ireader) in the reader coil needs to satisfy the saw tooth wave form. The reader can generate this current wave via the MCU and the operational amplifier. Another method is to modulate the current wave by pulse width or 4. Low Frequency Communication Implementation
4.1. The Frontend Circuit Design According to the low frequency data transmission model, the current (Ireader ) in the reader coil needs to satisfy the saw tooth wave form. The reader can generate this current wave via the MCU
Sensors 2017, 17, 867
5 of 14
and the operational amplifier. Another method is to modulate the current wave by pulse width or pulse density. The modulated signal is amplified by the operational amplifier, and sent by the Sensors 2017, 17, 867 5 of 14 reader. Figure 4 shows the current wave of the reader coil in pulse density modulation (PDM) mode. We compare the transmission distances the same card in three different modes: density. The modulated signal for is amplified by SIM the operational amplifier, and sent by the baseband Sensorspulse 2017, 17, 867 5 of 14 reader. Figure 4 shows the current wave of(PWM), the readerand coil in pulsewith density (PDM) mode. wave as transmission (BBT), pulse width modulation PDM a 4modulation MHz electromagnetic compare the transmission distances for the same SIM card in three different modes: baseband pulseWe density. modulated signal is amplified bywhich the operational amplifier, and sent by the the carrier. Our testThe results are introduced in Table 1, shows that the baseband transmission transmission (BBT), pulse width modulation (PWM), and PDM with a 4 MHz electromagnetic wave reader. Figure 4 shows the current wavetransmission of the reader coil in pulse density modulation (PDM) mode. and the PDM mode have about the same distance which is much longer than as the carrier. Our test results are introduced in Table 1, which shows that the baseband transmissionthat of the We compare the transmission distances for the same SIM card in three different modes: baseband PWM mode. micro-SIM has a larger outline dimension and antenna size than the nano-SIM, so the andThe the PDM mode have about the same transmission distance which is much longer than that of the transmission (BBT), width modulation and PDM with a size 4 MHz electromagnetic PWM mode. The micro-SIM has a larger dimension and antenna than the nano-SIM, so wave low frequency circuit ofpulse the micro-SIM hasoutline the(PWM), furthest working distance. When using other carrier as thethe carrier. Our test results are introduced Table 1, which shows distance. that the baseband transmission low frequency circuit of the micro-SIM in has the furthest working When using other frequencies, such as 2 MHz, 6 MHz, and 8 MHz, similar results are obtained. carrier frequencies, such as 2 MHz, 6 MHz, and 8 MHz,distance similar results are and the PDM mode have about the same transmission which isobtained. much longer than that of the PWM mode. The micro-SIM has a larger outline dimension and antenna size than the nano-SIM, so the low frequency circuit of the micro-SIM has the furthest working distance. When using other carrier frequencies, such as 2 MHz, 6 MHz, and 8 MHz, similar results are obtained.
Figure 4. The current wave of the reader coil in pulse density modulation mode.
Figure 4. The current wave of the reader coil in pulse density modulation mode. Table 1. Working distances of RFID SIM for three different modes in air.
Table 1.Distance Working(cm) distances BBT of RFID SIM for three different modes in air. PWM PDM Micro-SIM 13.5reader coil in 9.2 12.7 Figure 4. The current wave of the pulse density modulation mode. Distance (cm)12.0 BBT PWM PDM11.4 Nano-SIM 8.0 Table 1. Working distances of RFID SIM for three Micro-SIM 13.5 9.2 different 12.7 modes in air. The frontend circuit of the LF chip is one of the key components in low frequency Nano-SIM 12.0 8.0 11.4 communication. Before entering the signals needPDM to be processed in a BBTSIM LF chip,PWM Distance (cm) the RFID low-pass filter circuit, which is presented13.5 in Figure 5. The capacitors C1, C2, C3, C4, and the resistors Micro-SIM 9.2 12.7 R1 and R2 compose a filter circuit for filtering high-frequency noises. The values of the capacitors The frontend circuitNano-SIM of the LF chip is one of the key components in low frequency communication. 12.0 8.0 11.4 and the resistors need to be calculated and chosen discreetly when the system is in PWD and PDM Before entering the RFID SIM LF chip, the signals need to be processed in a low-pass filter circuit, mode. Besides the low frequency electromagnetic induction described by Equation (3), the resonance The frontend circuit of the LF chip is one of the key components in low frequency which is presented Figure The capacitors C1, C2, C3, C4, consists and the R1 antenna, and R2 compose effect of the in carrier is the5. key consideration. The resonance circuit of resistors low frequency communication. Before entering the RFID SIMThe LF chip, the need to and be processed in aneed and capacitors C1 and C2. The resonant frequency is described by a filter circuit for filtering high-frequency noises. values of signals the capacitors the resistors
filter circuit, which is presented in Figure C3, C4, and Besides the resistors 1 5. Theiscapacitors to be low-pass calculated and chosen discreetly when the=system in PWD C1, andC2, PDM mode. (4) the low 2π√ R1 and R2 compose a filter circuit for filtering high-frequency noises. The values of the capacitors frequency electromagnetic induction described by Equation (3), the resonance effect of the carrier is and the resistors need to be calculated discreetly whenLthe system is in PWD where f denotes a resonant frequency and equalchosen to the carrier frequency, is the inductance of the and low PDM the key consideration. The resonance circuit consists of low frequency antenna, and capacitors C1 and frequency antenna in the SIMelectromagnetic card, and C is the resonantdescribed capacitance load mode. Besides the low frequency induction by including Equation C1, (3), C2, the resonance C2. The resonant frequency is described by capacitance, andisparasitic capacitances. The The inductance of thecircuit low frequency the SIM card effect of the carrier the key consideration. resonance consistsantenna of low of frequency antenna, is about 5 toC1 10and nH, and the capacitance of C1 and can be calculated . C2 and capacitors C2. The resonant frequency is√ described by according to Equation (4).
f = 1 12π LC = 2π√
(4) (4)
wherewhere f denotes a resonant frequency carrierfrequency, frequency, L is inductance oflow the low f denotes a resonant frequencyequal equal to to the the carrier L is thethe inductance of the frequency antenna in the SIM card, and C is the resonant capacitance including C1, C2, load capacitance, frequency antenna in the SIM card, and C is the resonant capacitance including C1, C2, load and parasitic capacitances. The inductance ofinductance the low frequency ofantenna the SIMofcard is about capacitance, and parasitic capacitances. The of the lowantenna frequency the SIM card 5 to is and aboutthe 5 to 10 nH, and the capacitance of C1 C2 can be calculatedto according to (4). Equation (4). 10 nH, capacitance of C1 and C2 can beand calculated according Equation Figure 5. The front circuit of the low frequency chip.
Figure 5. 5. The lowfrequency frequency chip. Figure Thefront frontcircuit circuit of of the the low chip.
Sensors 2017, 17, 867
6 of 14
Sensors 2017, 17, 867
6 of 14
The filter circuit circuitconsists consistsofofthe theresistors resistors and capacitors C3 and C4. The low-pass low-pass filter R1R1 and R2,R2, andand capacitors C3 and C4. The The resistance of R1 and R2 cannot be too large in the low-pass filter circuit, which is described as resistance of R1 and R2 cannot be too large in the low-pass filter circuit, which is described as V0V0 == V VjwR1 (C3 + CL ) + 1 (5) (5) +1 1 3 + where where V0 V0 denotes denotes the the voltage voltage signal signal imported imported into into the the low low frequency frequency chip, chip, V V is is the the induced induced voltage voltage after after the the antenna antenna matching matching circuit, circuit, and and C CLL means the loading capacitance of the low frequency chip. After thethe optimized resistance of R1 30 to has this a lesshas negative Aftermany manyexperiments, experiments, optimized resistance ofand R1 R2 andis R2 is 50 30ohm; to 50this ohm; a less impact onimpact the weak induced negative on the weakvoltage. induced voltage. In In our our experiments, experiments, the the working working distance distance of low frequency frequency communication communication is affected distinctly by resistors and and capacitors capacitorsin inPDM PDMand andPWM PWMmodes. modes.Nevertheless, Nevertheless,this thissituation situation by the the values of the resistors is is notobserved observedininthe thebaseband basebandtransmission transmissionmode. mode.Based Basedon onthe the foregoing foregoing analysis analysis of the front not front circuit, and load modulation effects should alsoalso be considered in PDM and circuit, this thisimplies impliesthat thatresonance resonance and load modulation effects should be considered in PDM ◦ C, and PWM modes. a long conducting low temperature for approximately h 40 at −40 PWM modes. AfterAfter a long time time conducting low temperature teststests for approximately 72 h 72 at − we that the transmission distance decreases by 1.5~2.5 in PDM PWMand modes. However, °C,found we found that the transmission distance decreases by cm 1.5~2.5 cm and in PDM PWM modes. the communication distance for the same SIM card decreases by 0.5~1 cm in the baseband transmission However, the communication distance for the same SIM card decreases by 0.5~1 cm in the baseband mode. Therefore, the Therefore, baseband transmission is a bettermode scheme the 2scheme kHz LFfor communication transmission mode. the basebandmode transmission is afor better the 2 kHz LF in RFID SIM cardinapplications. communication RFID SIM card applications. 4.2. 4.2. Programmable Programmable Gain Gain Amplifiers Amplifiers of of the the LF LF Chip Chip The signal process process isisdescribed describedininFigure Figure6.6.The The weak signal (V0) from front circuit of The LF LF signal weak signal (V0) from thethe front circuit of the the low frequency chip needs to be amplified from tens of micro volts to a few volts. The three-stage low frequency chip needs to be amplified from tens of micro volts to a few volts. The three-stage programmable (PGA) andand oneone buffer are applied to dealtowith The amplified voltage programmablegain gainamplifiers amplifiers (PGA) buffer are applied dealit.with it. The amplified from the buffer and the output signal from the DAC are input into a comparator, which contributes to voltage from the buffer and the output signal from the DAC are input into a comparator, which the high and low level decision of the inductive signal. The circuit is also used to determine whether contributes to the high and low level decision of the inductive signal. The circuit is also used to adetermine card approaches departs, which in putswhich the 2.45 GHzputs chip in2.45 a waking up in or asleeping whether or a card approaches or turn departs, in turn the GHz chip waking mode accordingly. Thus, the bit stream produced by the comparator is decoded by the Manchester up or sleeping mode accordingly. Thus, the bit stream produced by the comparator is decoded by decoder [24]. Thedecoder decoded[24]. dataThe is checked bydata cyclicisredundancy algorithm. The(CRC) 8-bit the Manchester decoded checked bycheck cyclic(CRC) redundancy check 8 + x2 + x + 1. Lastly, an interrupt signal from the LF CRC circuit is designed with the polynomial x 8 2 algorithm. The 8-bit CRC circuit is designed with the polynomial x + x + x + 1. Lastly, an interrupt chip is from generated sent to the SE chip, and the SE SE chip accesses theSE LFchip chipaccesses and fetches data by and the signal the LFand chip is generated and sent to the chip, and the the LF chip serial peripheral interface. fetches data by the serial peripheral interface.
Figure 6. 6. The structure structure diagram diagram of of the the low low frequency frequency (LF) (LF) chip. chip. Figure
The PGA PGA and and buffer buffer circuits circuits are are described described in in Figure Figure 7.7. Each Each PGA PGA circuit circuit uses uses aa resistive resistive The feedback network. stage resistor network includes R1,R1, R21,R21, R22,R22, and and R23. R23. The feedback network. For Forexample, example,the thefirst first stage resistor network includes magnification is determined by the ratio of the feedback resistor to the gain resistor. The voltage gain The magnification is determined by the ratio of the feedback resistor to the gain resistor. The voltage can be by the configuration registers, which as gain cancontrolled be controlled by the configuration registers, whichchoose choosedifferent differentfeedback feedback resistors resistors such such as R21, R22, R22, and and R23. R23. The The second second and and third third PGA PGA has has the the same same circuitry. circuitry. Between Between each each stage stage of of PGAs, PGAs, R21, the low-pass filter circuit is carefully designed for removing high frequency noises, which contain the low-pass filter circuit is carefully designed for removing high frequency noises, which contain R7, R7, C1, R8, and C1, C3. and The C3. coupling The coupling capacitors C2,and C4, C5 andare C5applied are applied to block the direct current R8, capacitors C2, C4, to block the direct current level level signal. The signal V3 is inputted into the rail-to-rail output buffer, which enhances the load signal. The signal V3 is inputted into the rail-to-rail output buffer, which enhances the load capacity capacity of the PGA circuit. There are two voltage band-gap (BG) references in the LF chip. One of them is connected to the input ends of the antenna, the PGAs, and the buffer as 1 V common-mode
Sensors 2017, 17, 867
7 of 14
of the PGA circuit. There are two voltage band-gap (BG) references in the LF chip. One of them is Sensors 2017, 17, 867 7 of 14 connected to the input ends of the antenna, the PGAs, and the buffer as 1 V common-mode voltage in Sensors 2017,1.2 17, 867 7 of 14 Figure reference which is used to generate thegenerate digital and voltage7.inThe FigureV7.BG Theworks 1.2 V as BGthe works as thevoltage, reference voltage, which is used to the analog digital voltage sources. and analog voltage sources. voltage in Figure 7. The 1.2 V BG works as the reference voltage, which is used to generate the digital and analog voltage sources.
Figure gainamplifiers amplifiers circuit. Figure7.7.Programmable Programmable gain circuit.
Figure 8 describes the loop gainphase phase and the loop gain frequency response of the We We Figure 88 describes thethe loop gain frequency response of PGAs. the Figure describes the theloop loopgain gain phaseand and loop gain frequency response of PGAs. the PGAs. simulated the PGA circuits at −40 °C◦, 27 °C◦, and 85 °C ◦for different process corners including slow, °C, C, and 8585 °C C forfor different simulated thethe PGA circuits at at −40−°C We simulated PGA circuits 40 , 27 C, 27 and differentprocess processcorners cornersincluding includingslow, slow, fast, typical, slow N fast P, and fast N slow P models. Therefore, all combinations result in 15 curves typical, slow slowN Nfast fastP,P,and andfast fastNNslow slowPPmodels. models.Therefore, Therefore, all combinations result in 15 curves fast, typical, all combinations result in 15 curves in in Figure 8a and Figure 8b. These curves show that the proposed PGA circuits work correctly in in Figure 8a and Figure 8b. These curves show that the proposed PGA circuits work correctly in Figure 8a,b. These curves show the proposed PGAachieve circuitsthe work correctly inofdifferent temperatures different temperatures and that process corners, and requirement the small signal different temperatures process corners, and achieve requirement of atthe small signal and process corners, achieve the requirement of the small the signal amplification 2 kHz. amplification at 2 and kHz.and amplification at 2 kHz.
(a)
(a)
(b) Figure 8. (a) Loop gain phase frequency response of programmable gain amplifiers (PGAs) at the
Figure 8. (a) Loop gain phase frequency response of programmable gain amplifiers (PGAs) at the corners of −40 °C, 27 °C, and 85 °C; (b) Loop gain frequency response of PGAs at the corners of −40 °C, corners of −40 ◦ C, 27 ◦ C, and 85 ◦ C; (b) Loop gain frequency response of PGAs at the corners of 27 °C, and 85 °C. (b) −40 ◦ C, 27 ◦ C, and 85 ◦ C.
Figure 8. (a) Loop gain phase frequency response of programmable gain amplifiers (PGAs) at the corners of −40 °C, 27 °C, and 85 °C; (b) Loop gain frequency response of PGAs at the corners of −40 °C, 27 °C, and 85 °C.
Sensors 2017, 17, 867
8 of 14
4.3. Differential Manchester Decoder A Differential Manchester code is a line code in which data and clock signals are combined to form a single 2-level self-synchronizing data stream. In the Differential Manchester code stream, Sensors 2017, 17, 867 8 of 14 0 is represented by no transition and 1 is represented by a transition [25,26]. The Manchester decoder 4.3. Differential Manchester Decoder needs to estimate the long and the short signal level to judge the transition. The common method of Manchester decoding is that the signal between two transitions is signals sampled and recorded by the A Differential Manchester codelevel is a line code in which data and clock are combined to form a single 2-level self-synchronizing data stream. In the Differential Manchester code stream, 0 is sampling clock. Then, the decoder compares the number of samples and the threshold value of the represented by no transition and 1 is represented by a transition [25,26]. The Manchester decoder signal length to distinguish between the long and short signal levels. Lastly, the symbols 0 and 1 are needs to estimate the long and the short signal level to judge the transition. The common method of exported by a decoder. Accordingly, sampling clock important to decode Manchester decoding is that the the signal level between twoprecision transitionsisis very sampled and recorded by the the code stream correctly, methodstheare proposed [27,28]. A threshold serious value problem samplingand clock.some Then, anti-noise the decoder compares number of samples and the of theis that the signal length to distinguish between circuit the longused and short signal levels. the symbols 0 andof 1 are clocks generated by the linear oscillator a resistive andLastly, capacitive network the different exported by a decoder. Accordingly, the sampling clock precision is very important to decode the LF chips are unstable and that different chips have different frequencies. In the worst case, the clock code stream correctly, and some anti-noise methods are proposed [27,28]. A serious problem is that frequencythe deviation reachesbyfrom −20% to +20%, and this resultsand in acapacitive higher network Manchester clocks generated the linear oscillator circuit used a resistive of the decoding error rate. different LF chips are unstable and that different chips have different frequencies. In the worst case, the clock frequency deviation reaches from −20% to +20%, and is this results in ain higher Manchester The proposed differential Manchester decoding method illustrated Figure 9. The decoding decoding error rate. signal is first filtered by the filtering circuit to remove the glitches. In the sample counter circuit, The proposed differential Manchester decoding method is illustrated in Figure 9. The decoding when catching high the counter 1; when sampling the low level signal, signal is first level filteredsignals, by the filtering circuit toincreases remove theby glitches. In the sample counter circuit, the counter decreases by 1.level Consequently, the increases square waves aresampling transformed intosignal, triangular waves, when catching high signals, the counter by 1; when the low level the counter decreases byThe 1. Consequently, the square waves are transformed into triangular into waves, as as illustrated in Figure 10. high and low level threshold values are imported a comparator, illustrated in Figure 10. The high and low level threshold values are imported into a comparator, which is similar to the function of the Schmitt trigger. When the triangular waves are greater than the which is similar to the function of the Schmitt trigger. When the triangular waves are greater than high threshold, thethreshold, intermediate data is 1;data when triangular waves are are lessless than the high the intermediate is 1;the when the triangular waves thanthe thelow low threshold, the intermediate data 0. The transition is used to generate decoding clock.clock. There are two threshold, the is intermediate data is 0. detector The transition detector is used to the generate the decoding are two conditions to generate a decoding One is that the positive trigger occurs conditionsThere to generate a decoding clock. One is thatclock. the positive edge triggeredge occurs and the triangular and the triangular waves are less than the low level threshold, the other is that the negative edge waves are less than the low level threshold, the other is that the negative edge trigger occurs and the trigger occurs and the triangular waves are greater than the high level threshold. Then, the decoding triangularclock waves are greater the level which threshold. the decoding clock is fed into the is fed into the twothan groups of high D flip-flops, performThen, the two stages of the intermediated two groups ofsampling. D flip-flops, perform the two stages the intermediated Lastly, data Lastly,which the decoded data is generated by the of XOR operation of the two data stagessampling. of the D flip-flops. the decoded data is generated by the XOR operation of the two stages of the D flip-flops.
9. architecture The architectureofofthe the differential Manchester decoder. FigureFigure 9. The differential Manchester decoder.
Three-fourths of the maximum value from the sampling counter is used as the high level
Three-fourths of the maximum from thelow sampling counter is used as the high edges, level threshold, threshold, and a quarter of thatvalue is applied as the level threshold. When catching negative the value of the imported record the register. SE chipnegative reads values many the value and a quarter of that is sampling appliedcounter as theislow leveltothreshold. When The catching edges, times and calculates the mean value as the maximum value. Therefore, the SE chip sets the default of the sampling counter is imported to record the register. The SE chip reads values many times and value as the maximum according to the original clock frequency, and the value of the high and low calculates the mean value as the maximum value. Therefore, the SE chip sets the default value as the maximum according to the original clock frequency, and the value of the high and low level thresholds can be obtained adaptively according to the real clock frequency. When adopting triangular waves, the long and short signal levels are not divided by the fixed threshold values. The clock skew only
Sensors 2017, 17, 867 Sensors 2017, 17, 867 Sensors 2017, 17, 867
9 of 14 9 of 14 9 of 14
level level thresholds thresholds can can be be obtained obtained adaptively adaptively according according to to the the real real clock clock frequency. frequency. When When adopting adopting triangular triangular waves, waves, the the long long and and short short signal signal levels levels are are not not divided divided by by the the fixed fixed threshold threshold values. values. The The causes a difference in the height of the wave. After using the After adaptive high and low level threshold clock clock skew skew only only causes causes aa difference difference in in the the height height of of the the wave. wave. After using using the the adaptive adaptive high high and and low low scheme, the effects caused by the clock frequency deviation are eliminated. level threshold scheme, the effects caused by the clock frequency deviation are eliminated. level threshold scheme, the effects caused by the clock frequency deviation are eliminated. Reference Reference Input stream Input stream
Triangular Triangular waves waves
High level High level
Low level Low level Decoding clock Decoding clock
Intermediate data Intermediate data Output stream Output stream
1 1
1 1 0 0
1 1
1 1 0 0
0 0
Figure 10. The Differential Manchester decoding process. Figure 10. The The Differential Differential Manchester decoding process.
We added some test pins for the decoder. The test between the We addedsome some for testing testing the Manchester Manchester Thebetween test pin pinthe between the We added testtest pinspins for testing the Manchester decoder.decoder. The test pin comparator comparator and the Manchester decoder in Figure 6 is added and connected to a field programmable comparator and the Manchester decoder in Figure 6 is added and connected to a field programmable and the Manchester decoder in Figure 6 is added and connected to a field programmable logic arrays logic arrays (FPGA) board. The digital logic including the Manchester decoder, the data fist logic arrays (FPGA) board.logic The chips digitalincluding logic chips chips thedecoder, Manchester decoder, fist in in (FPGA) board. The digital theincluding Manchester the data fist inthe fistdata out (FIFO) fist out (FIFO) modules, the CRC circuit, and the SPI circuit are implemented by the FPGA. The fist out (FIFO) modules, the CRC circuit, and the SPI circuit are implemented by the FPGA. The modules, the CRC circuit, and the SPI circuit are implemented by the FPGA. The expected digital expected digital frequency is 380 kHz, and test frequency is from 270 kHz to 490 kHz expected digital clock clock 380test kHz, and our ouris test frequency from 490 intervals. kHz at at 11 clock frequency is 380frequency kHz, and is our frequency from 270 kHzisto 490270 kHzkHz at 1tokHz kHz intervals. The clock is generated and controlled by a digital function generator, which is kHz intervals. The clock generated controlled a digitalwhich function generator, which is The clock is generated and is controlled by aand digital functionby generator, is connected to the FPGA connected to the FPGA as the decoding clock. The experiment result shows that this circuit can connected to the FPGA as the decoding clock. The experiment result shows that this circuit can as the decoding clock. The experiment result shows that this circuit can decode successfully from decode successfully from 270 kHz 490 kHz, means differential Manchester decoder can decode 270 kHz to to 490 kHz, which which means this thisdecoder differential Manchester can 270 kHzsuccessfully to 490 kHz,from which means this differential Manchester can work under adecoder clock skew work under a clock skew of approximately ±30%. work under a clock skew of approximately ±30%. of approximately ±30%. 5. 5. Implementation Implementation and and Experimental Experimental Results Results The The LF LF chip chip has has been been fully fully implemented implemented in in Central Central Semiconductor Semiconductor Manufacturing Manufacturing Corporation Corporation (CSMC) 0.18 µm mixed signal technology with one layer of polycrystalline (CSMC) 0.18 µm mixed signal technology with one layer of polycrystalline silicon silicon and and five five layers layers of of metal. The die size is 1850 µ m by 1650 µ m with a 330 µ A working current and a 45 µ A idle current, metal. The m by m with A working A idle The die die size size is is 1850 1850 µµm by 1650 1650 µ µm with aa 330 330 µ µA working current current and and aa 45 45 µµA idle current, current, which which can can satisfy satisfy the the current current requirements requirements of of the the SIM SIM card. card. Figure Figure 11 11 gives gives the the low low frequency frequency chip chip layout and photograph. The low frequency chip includes six DACs and comparators for different layout and photograph. The The low low frequency frequency chip chip includes includes six six DACs DACs and comparators comparators for different functions, and departure detection. functions, such as the Manchester code stream generation, and approach functions, such such as asthe theManchester Manchestercode codestream streamgeneration, generation,and andapproach approachand anddeparture departuredetection. detection. Figure 11b also shows that there are some current hot spots near the band-gap and LDO Figure 11b areare some current hothot spots nearnear the band-gap and LDO in the Figure 11balso alsoshows showsthat thatthere there some current spots the band-gap and chips LDO chips chips in the low frequency chip. low frequency chip. chip. in the low frequency
(a) (a)
(b) (b)
Figure 11. The layout and photograph of the low frequency chip (a) layout; (b) photograph. Figure 11. The layout and photograph of the low frequency chip (a) layout; (b) photograph. Figure 11. The layout and photograph of the low frequency chip (a) layout; (b) photograph.
Sensors 2017, 17, 867 Sensors 2017, 17, 867
10 of 14 10 of 14
Figure 12 shows the test platform of the LF chip and 2.45 GHz chips. To guarantee testing availability and reliability, the LF chip uses the coils from the real RFID SIM card substrate, Sensors 2017, 867 10 ofand 14 the Figure 1217, shows the test platform of the LF chip and 2.45 GHz chips. To guarantee testing reader is from the commercial products. The LF chip with the chip on board (COB) package is fitted availability and 12 reliability, the LFplatform chip uses coilschip from the real RFID SIMTo card substrate, and the Figure shows the test of the the and 2.45 GHz chips. guarantee testing into the test motherboard. The SIM substrate in LF the upper left corner consists of low frequency coils reader is from the commercial products. Thethe LFcoils chipfrom withthe thereal chip on SIM board (COB) package is fitted availability and reliability, the LFconnected chip uses RFID card substrate, and the and the frontend circuit, which is to the LF chip and the motherboard via wires. The ISO into the test motherboard. Theproducts. SIM substrate the upper lefton corner of lowis frequency reader is from the commercial The LF in chip with the chip board consists (COB) package fitted 7816 interface is in the top right corner of the motherboard, and transfers data from the debug host. motherboard. Thewhich SIM substrate in the upper low frequency via coilswires. coils into andthe thetest frontend circuit, is connected to theleft LFcorner chip consists and theofmotherboard The debugging board is at the top, which provides power and the debugging interface for the and the frontend circuit, which is connected to the LF chip and the motherboard via wires. The ISO The ISO 7816 interface is in the top right corner of the motherboard, and transfers data from the debug motherboard. The dual-mode reader is on the right-hand side. The same debugging board is is inboard the topisright corner the motherboard, and transfers from the interface debug host. host.7816 The interface debugging at the top,of which provides power and thedata debugging for the connected to the reader. right is the antenna board reader, interface which contains The debugging board To is atthe thefar top, which provides power and for the the debugging the low motherboard. The dual-mode reader is on the right-hand side. The same debugging board isforconnected frequency loop coils and a 2.45 GHz high frequency rod antenna. motherboard. The dual-mode reader is on the right-hand side. The same debugging board is to the reader. To the far right is the antenna board for the reader, which contains low frequency loop connected to the reader. To the far right is the antenna board for the reader, which contains low coils and a 2.45 GHz high frequency rod antenna. frequency loop coils and a 2.45 GHz high frequency rod antenna.
Figure 12.12. The testboard boardofof the RFID cards. Figure Thelow lowfrequency frequency chip chip test the RFID SIMSIM cards. Figure 12. The low frequency chip test board of the RFID SIM cards.
We tested communicationdistance distance and and waveforms V3V3 signal in Figure 7 for7the We tested thethe communication waveformsofofthethe signal in Figure forLF the LF chip using the test platform. Figure 13 illustrates the waveforms of V3 for different distances tested communication distance and waveforms of the V3 signal in Figure 7 for thedistances LF chip chip We using thethe test platform. Figure 13 illustrates the waveforms of V3 for different between the reader antenna and the LF coil the in the SIM card substrate. waveform in Figure13a is the using thethe testreader platform. Figure 13the illustrates waveforms V3 forThe different distances between antenna and LF coil in the SIM card of substrate. The waveform in between Figure13a is consistent with V SIM in Figure 3 described by the low frequency transmission model. Figure 13b reader antenna coil in the SIM card The waveform in Figure 13a isFigure consistent consistent with and VSIMthe in LF Figure 3 described by substrate. the low frequency transmission model. 13b shows that the wave distortion of V3 causes incorrect Manchester decoding in the critical region. In with VSIM inthe Figure describedofbyV3the low frequency transmission model. in Figure 13b shows that shows wave3distortion causes Manchester decoding the critical region. In ourthat tested experiments, the mean error ratesincorrect are measured by the CRC error interrupt. The mean the wave distortion of V3 causes incorrect Manchester decoding in the critical region. In our tested our tested experiments, the mean error rates are measured by the CRC error interrupt. The mean error rate is lower than 2% when the LF chip is in the normal decoding region. However, the mean experiments, the mean error ratesthe are measured bythe theregion, CRC error interrupt. TheHowever, mean error is errorerror rate rate is lower than 2% beyond when chipdecoding is in normal decoding region. the mean grows quickly theLF critical and decoding failures will cause therate lower than 2% when the LF chip is in the normal decoding region. However, the mean error rate grows errorwhole rate grows quickly beyond thethis, critical decoding region, and decoding cause the application to fail. Besides our test results also demonstrate that thefailures chip haswill a better quickly beyond the to critical region, and failures willentering cause the whole application to communication distance and decoding whenalso the demonstrate PGAs into the whole application fail. decoding Besides this, ourstability test decoding results that the chipsaturated has a better amplification mode. fail. Besides this, distance our test results also demonstrate the chip a better communication distance communication and decoding stabilitythat when the has PGAs entering into the saturated
and decoding mode. stability when the PGAs entering into the saturated amplification mode. amplification
(a)
(b)
Figure 13. (a) Waveform of the normal decoding region; (b) Waveform of the critical decoding (a) (b) region.
Figure 13. (a) Waveform of the normal decoding region; (b) Waveform of the critical decoding Figure 13. (a) Waveform of the normal decoding region; (b) Waveform of the critical decoding region. region.
Sensors 2017, 17, 867
11 of 14
Sensors 2017, 17, 867
11 of 14
The The micro micro and and nano-SIM nano-SIM cards cards are are in in the the lower lower right right corner corner of of Figure Figure 12, 12, which which show show the the same same shape as the ordinary SIM cards. To integrate the secure element chip and low frequency chip, the back shape as the ordinary SIM cards. To integrate the secure element chip and low frequency chip, the thinned process has been the low frequency circuit wafer andwafer the SEand wafer purpose of back thinned process hasdone beenfor done for the low frequency circuit thefor SEthe wafer for the stacking. The RFID SIM cards can fit almost all mobile phones. To ensure the adaptability of working purpose of stacking. The RFID SIM cards can fit almost all mobile phones. To ensure the adaptability ◦ C, 0 ◦ C, 27 ◦ C, 60 ◦ C, and 85 ◦ C for conditions, performed tests with −tests 40 ◦ C, −20−40 of working we conditions, we temperature performed temperature with °C, −20 °C, 0 °C, 27 °C, 60 °C, and the RFID SIMRFID cards.SIM Thecards. temperature test platform is shown is inshown Figurein 14Figure including a high and low 85 °C for the The temperature test platform 14 including a high temperature chamber. We chose different groups of RFID SIM cards and placed them in the chamber and low temperature chamber. We chose different groups of RFID SIM cards and placed them in the for 72 h atfor each Then the different groups of RFID SIM cards are tested for communication chamber 72temperature. h at each temperature. Then the different groups of RFID SIM cards are tested for distances respectively in the air, and by using the test board of the RFID SIM cards. The communication distances respectively in the air, and by using the test board of the RFID test SIMresults cards. are illustrated in Table 2, which shows the working distances in the baseband transmission, PWM, The test results are illustrated in Table 2, which shows the working distances in the baseband and PDM modes for different temperatures. The experiment reveals that basebandreveals transmission is transmission, PWM, and PDM modes for different temperatures. The the experiment that the abaseband better working distance and more excellent temperature adaptability, and that high temperatures transmission is a better working distance and more excellent temperature adaptability, have less high influence on the working distances of RFID cards than low temperatures. The RFID and that temperatures have less influence on theSIM working distances of RFID SIM cards than SIM low cards in the BBTThe mode show good performance at almost all the good temperatures, but there exists aall small temperatures. RFID SIM cards in the BBT mode show performance at almost the ◦ C. decrease for communication distances at − 40 temperatures, but there exists a small decrease for communication distances at −40 °C.
Figure 14. Temperature test chamber for RFID SIM cards. Figure 14. Temperature test chamber for RFID SIM cards. Table 2. Working distances of nano-SIM cards at different temperatures in air for the baseband Table 2. Working distances of nano-SIM cards at different temperatures in air for the baseband mode. mode. ◦ −40 ◦ C −40 °C−20−20C°C
Distance BBT (cm) 11.2 11.2 Distance in BBTin (cm) DistanceDistance in PWMin(cm) PWM(cm)6.5 6.5 Distance in PDMin(cm) Distance PDM (cm)9.2 9.2
11.811.8 7.9 7.9 11.211.2
00◦°C C 12.0 12.0 7.9 7.9 11.3 11.3
◦ 2727 °C C 12.0 12.0 8.08.0 11.4 11.4
60 ◦ C 85 °C 85 ◦ C 60 °C 12.0 12.0 12.0 12.0 7.9 7.9 7.9 7.9 11.3 11.3 11.3 11.3
To study study the the effects effects of of low low temperature temperature for for the the LF LF communication, communication, we we used used 10 10 COB COB LF LF chips chips To and SIM substrates with antennas to do some experiments. Firstly, the electrical parameters of the LF and SIM substrates with antennas to do some experiments. Firstly, the electrical parameters of the LF ◦ chips are tested at 27 °C, and depicted in Table 3. The parameters are measured from the testing pins chips are tested at 27 C, and depicted in Table 3. The parameters are measured from the testing pins of COB COB chips, chips, including including digital digital device device voltages voltages (D_VDD), (D_VDD), analog analog device device voltages voltages (A_VDD), (A_VDD), voltage voltage of source references from BG (V_BG), PGAs’ common-mode voltages from BG (P_BG), the frequencies source references from BG (V_BG), PGAs’ common-mode voltages from BG (P_BG), the frequencies of internal internal oscillators, oscillators, and and the the working working current current of of the the LF LF chips. chips. Then, Then, the the chips chips and and substrates substrates are are of ◦ placed into intoaatemperature temperaturechamber chamberfor for h at −40C. °C.The The chips tested again results placed 7272 h at −40 LFLF chips areare tested again andand the the results are are shown in Table 4. Apart from the working current, all the variations of the electrical parameters shown in Table 4. Apart from the working current, all the variations of the electrical parameters are no are no more than from Tables 3 mean and 4.working The mean working current increased by 2.7% and more than 0.3% from0.3% Tables 3 and 4. The current increased by 2.7% and communication communication decreased by 6.7% for BBT mode, which shows there is a possible correlation decreased by 6.7% for BBT mode, which shows there is a possible correlation between the variation of between the variation of the working current and that of the distance. Besides this, when we changed a new SIM substrate with the front circuit and LF coils without staying at the low
Sensors 2017, 17, 867
12 of 14
the working current and that of the distance. Besides this, when we changed a new SIM substrate with the front circuit and LF coils without staying at the low temperature, the working distance showed some improvements for PWM and PDM modes, but very little change for BBT modes. Therefore, the low temperature at −40 ◦ C will bring about many adverse influences on the LF transmission, first of which is the values of capacitors and resistors in the front circuit change, which causes a decrease in the working distance in PDM and PWM modes. Secondly, the variations of the working current imply a leakage current increase of the LF chips. Lastly, some changes of the circuit parameters in the LF chip and increased noise by leakage current possibly affect the amplification quality for the weak signal. All these lead to working distance degradation at low temperature after a long period. Table 3. Electrical parameters of the LF chips at 27 ◦ C. No.
D_VDD (V)
A_VDD (V)
V_BG (V)
P_BG (V)
OSC (kHz)
Current (µA)
1 2 3 4 5 6 7 8 9 10
1.799 1.844 1.785 1.838 1.850 1.801 1.857 1.819 1.856 1.824
1.960 1.996 1.954 2.000 1.975 1.943 2.000 1.967 2.000 1.969
1.150 1.165 1.144 1.167 1.169 1.134 1.173 1.161 1.171 1.159
0.989 0.974 0.991 1.007 1.019 1.020 1.009 1.008 0.982 0.989
402.17 391.02 391.74 402.4 376.94 372.33 388.56 376.56 355.17 382.43
325.5 329.1 328.5 335.2 333.4 327.3 229.7 324.3 341.2 325.5
Table 4. Electrical parameters of the LF chips after 72 h at −40 ◦ C. No.
D_VDD (V)
A_VDD (V)
V_BG (V)
P_BG (V)
OSC (kHz)
Current (µA)
1 2 3 4 5 6 7 8 9 10
1.801 1.848 1.782 1.846 1.854 1.805 1.864 1.824 1.859 1.820
1.963 1.994 1.955 2 1.977 1.947 2 1.966 2 1.963
1.151 1.167 1.147 1.184 1.172 1.14 1.176 1.163 1.172 1.155
0.99 0.974 0.993 1.007 1.02 1.024 1.01 1.011 0.982 0.986
401.95 391.3 391.53 402.67 377.61 372.35 386.24 376.55 356.58 381.56
331.5 340 330.4 339.4 338.8 331 335.8 328.5 347.8 330.4
We installed the dual-mode reader into the retail POS (point of sale) terminal of the Newcapec Company, and compared the communication distance and success ratio of Manchester decoding for the RFID SIM card at room temperature in air and in different mobile phones, including Apple, Huawei, and Samsung. The test results are summarized in Table 5, which shows that the LF chips have excellent near field communication ability and can work within a 7 cm range for almost all popular mobile phones. The NFC works at distance about 6 cm for the Apple iPhone 6S with the Newcapec POS terminal. SIM cards in air have the longer working distance and the higher decoding success ratio. The metal shell shielding of mobile phones has more influence on the working distance and the decoding success ratio than plastic shells. The experiments also show that if the decoding success ratio is more than 80%, the payment application can work correctly because the LF reader sends the same waking up and configuration data repeatedly.
Sensors 2017, 17, 867
13 of 14
Table 5. Working distances and Manchester decode success ratio of RFID SIM for different mobile phones and in air.
Distance (cm) Success ratio
Iphone 6
Iphone 5s
Huawei Mate 7
Samsung I9200
In Air
7.8 93%
8.2 95%
7.5 91%
8.8 98%
12 99%
6. Conclusions The integration of the 2.45 GHz RFID chip and the LF chip for RFID SIM cards suitable for middle and long distance RFID applications and near field payments has been discussed and implemented. Compared to ordinary SIM and e-SIM cards, RFID SIM cards consist of a secure element chip, a 2.45 GHz RF chip, and a low power range controlled communication chip with a 2 kHz working frequency. This RFID SIM can support both long distance and near distance applications with reasonable power consumption. For future work, there are some issues worthy of further study in the LF chip design, for example the mechanism for working distance degradation at low temperature; the metal shielding effect of mobile phone shells; the relationship between the working distance and the direction of operation; the reduction of communication crosstalk between a low frequency and 13.56 MHz; and the trade-off between power consumption and working distance at different operating frequencies (e.g., 4 kHz, 8 kHz, and 10 kHz). Acknowledgments: The work is partially supported by the Shenzhen Key Lab for RF Integrated Circuits, the Shenzhen Shared Technology Service Center for Internet of Things, the Technology Innovation Fund for China Technology Enterprise (No. 14C26214422753), the Major Program of Key Technology R&D of Hubei Province of China (Grant No. 2015ACA063), the National Natural Science Foundation of China (Grant number 61006020), the Fundamental Research Funds for the Central Universities (HUST: 2014TS041), National key R&D plan (grant number 2016YFC0105002 and 61674162), Guangdong government funds (grant numbers 2013S046 and 2015B010104005), Shenzhen government funds (grant numbers CXZZ20150601160410510 and JCYJ20160331192843950), and Shenzhen Peacock Plan. Author Contributions: Yicheng Chen and Fengqi Yu contributed to the main idea of this paper; Zhaoxia Zheng conceived and designed the experiments; Mingyang Gong performed the experiments; Yicheng Chen and Mingyang Gong analyzed the data; Yicheng Chen wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3.
4. 5. 6.
7.
8.
Want, R. An Introduction to RFID Technology. IEEE Pervasive Comput. 2010, 5, 25–33. [CrossRef] Tan, W.H.; Ooi, K.B.; Chong, S.C.; Hew, T.S. NFC mobile credit card: the next frontier of mobile payment? Telemat. Inform. 2014, 31, 292–307. [CrossRef] Pasquet, M.; Reynaud, J.; Rosenberger, C. Secure Payment with NFC Mobile Phone in the Smart Touch Project. In Proceedings of the International Symposium on Collaborative Technologies and Systems, Irvine, CA, USA, 19–23 May 2008; pp. 19–23. Bolhuis, M. Using an NFC-Equipped Mobile Phone as a Token in Physical Access Control. Available online: http://essay.utwente.nl/65419/1/thesis_nfc_martijn_bolhuis_final.pdf (accessed on 3 May 2014). Liu, F.; Janssens, D.; Wets, G.; Cools, M. Annotating mobile phone location data with activity purposes using machine learning algorithms. Expert Syst. Appl. 2013, 8, 3299–3311. [CrossRef] Henpraserttae, A.; Thiemjarus, S.; Marukatat, S. Accurate Activity Recognition Using a Mobile Phone Regardless of Device Orientation and Location. In Proceedings of the 2011 International Conference on Body Sensor Networks, Dallas, TX, USA, 23–25 May 2011; pp. 23–25. Yang, G.; Huang, Z.; Wan, L. The development of RFID module in NFC phone. In Proceedings of the International Conference on Anti-Counterfeiting, Security, and Identification in Communication, Hong Kong, China, 20–22 August 2009. Von Oheimb, D.; Lotz, V.; Walter, G. Analyzing SLE 88 memory management security using Interacting State Machines. Int. J. Inf. Secur. 2005, 4, 155–171. [CrossRef]
Sensors 2017, 17, 867
9. 10. 11.
12. 13. 14.
15. 16.
17.
18. 19. 20.
21. 22. 23. 24. 25. 26. 27. 28.
14 of 14
Hsu, C.C.; Chen, J.H. A novel sensor-assisted rfid-based indoor tracking system for the elderly living alone. Sensors 2011, 11, 10094–10113. [CrossRef] [PubMed] Kim, S.H.; Park, H.; Bang, H.C.; Kim, D.H. An indoor location tracking based on mobile rfid for smart exhibition service. J. Comput. Virol. Hacking Tech. 2014, 2, 89–96. [CrossRef] Ni, L.M.; Liu, Y.; Lau, Y.C.; Patil, A.P. LANDMARC: Indoor location sensing using active RFID. In Proceedings of the IEEE International Conference on Pervasive Computing and Communications, Fort Worth, TX, USA, 23–26 March 2003. Qing, X.; Chen, Z.N. Proximity Effects of Metallic Environments on High Frequency RFID Reader Antenna: Study and Applications. IEEE Trans. Antennas Propag. 2007, 55, 3105–3111. [CrossRef] Lee, B.; Kim, B.; Mum, B.; Lee, H. NFC Antenna Design for Low-permeability Ferromagnetic Material. IEEE Antennas Wirel. Propag. Lett. 2014, 13, 59–62. Yu, S.; Peng, Y.; Huang, X. Realization of Coordinative Control between Multi Readers and Multi RF-SIM Cards under Mobile RF-SIM Mode. In Proceedings of the International Conference on Computer Science and its Applications, Jeju, Korea, 22–25 November 2012. Ying, L.I.; Hong, L.I.; Chen, Y.T. Design on highway charge system based on RFSIM path recognition. Mod. Electron. Tech. 2014, 37, 60–63. Dai, W.; Luo, G.; Liu, S. A new technology of short-range communication for one card solutionin school. In Proceedings of the IET International Conference on Communication Technology and Application, Beijing, China, 14–16 October 2011. Wei, Q.; Wang, X. The Design and Implementation of the Minimal M2M Terminal Based on the SIM4100 Wireless Module. In Proceedings of the Third International Symposium on Information Processing, Qingdao, China, 15–17 October 2010. Shan, L.; Li, Z.; Hu, H. Converged Mobile Cellular Networks and Wireless Sensor Networks for Machine-to-Machine Communications. Ksii Trans. Internet Inf. Syst. 2012, 1, 147–161. [CrossRef] Smart Cards Embedded UICC Requirements Specification (Release 13): ETSI TS 103.383. Available online: http: //www.etsi.org/index.php/technologies-clusters/technologies/smart-cards (accessed on 1 September 2013). Remote Provisioning Architecture for Embedded UICC Technical Specification Version 3.1: GSMA. SGP.02. Available online: http://www.gsma.com/newsroom/all-documents/sgp-02-v3-1-remote-provisioningarchitecture-for-embedded-uicc-technical-specification/ (accessed on 27 May 2016). RSP Technical Specification Version 1.1: GSMA. SGP.2. Available online: http://www.gsma.com/newsroom/ all-documents/sgp-22-technical-specification-v1-1/ (accessed on 9 September 2016). Understanding SIM Evolution: GSMA. Available online: https://www.gsmaintelligence.com/research/ 2015/03/understanding-sim-evolution/499/ (accessed on 3 March 2015). Klaus, F.; Müller, D. RFID Handbook: Fundamentals and Applications in Contactless Smart Cards, Radio Frequency Identification and Near-Field Communication, 3rd ed.; Wiley: Hoboken, NJ, USA, 2010; pp. 35–55. Chen, Y.; Zheng, Z. A low-power 2K/4K range-controlled communication chip design for mobile payment. In Proceedings of the IEEE International Nanoelectronics Conference, Chengdu, China, 9–12 April 2016. Hung, Y.C. Time-interleaved CMOS chip design of Manchester and Miller encoder for RFID application. Analog. Integr. Circuits Signal Process. 2012, 71, 549–560. [CrossRef] Saltzberg, B. An improved Manchester code receiver. In Proceedings of the IEEE International Conference on Communications, Atlanta, GA, USA, 16–19 April 1990. Guo, X.; Chen, Z. A New Circuit for Manchester Decoder Based on FPGA. Open J. Circuits Syst. 2014, 3, 31–36. [CrossRef] Xu, D.C.; Lu, J.F. Design of Manchester codec based on FPGA. Adv. Mater. Res. 2012, 614, 1357–1360. [CrossRef] © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
