2014 5th International Conference on Computer and Communication Technology
A New LEACH-based Routing Protocol for Energy Optimization in Wireless Sensor Network Jyoti Singh1
Bhanu Pratap Singh1, Subhadra Shaw2
Department of Computer science Vindhya Institute of Technology & Science1 Satna, Madhya Pradesh, India E-mail:
[email protected]
E-mail:
[email protected] CSED, MNNIT, Allahabad, India2 E-mail:
[email protected] networks must be managed wisely to extend the lifetime of sensors. 70% of the energy is consumed in transmission of data between sensor nodes and base station [8]. In addition, it is very important to balance the energy consumption among all sensor nodes to prolong the network lifetime.
Abstract— Due to lack of power source, energy is the key concern area in Wireless Sensor Networks. Maximum energy is used in transmission of data. Many research works have been done to develop routing algorithms to increase the lifetime of a sensor network. Among them clustering based approach is well known for achieving energy efficiency. This paper proposes a new routing strategy based on hierarchical routing protocol LEACH where clusters are refreshed periodically based on residual energy and distance. Reclustering distributes the workload among different nodes and in turn enhances the network lifetime by rotating the cluster head. The sensor nodes remain in active state only during its transmission slot. Rest of the time it remains in sleep state to save energy. LEACH, MOD-LEACH and the proposed protocol are simulated in MATLAB. The result shows that our proposed algorithm performs better than the LEACH and also MOD-LEACH protocol in terms of network lifetime. The proposed algorithm also gives more throughput than LEACH.
In order to achieve high energy efficiency and increase the network scalability clustering routing protocol can be used. In this type routing mechanism sensor nodes are grouped into clusters. Sensors sense data from the environment and send it to the cluster head. Since individual node’s data are correlated in a micro-sensor network, the end-user does not require all the (redundant) data; rather, he/she only needs a high-level function of the data that describes the events occurring in the environment. So the cluster head aggregates the data obtained from all the cluster members before forwarding it to the base station. Data aggregation helps to minimize data redundancy and reduce the communication load [2] because much less actual data needs to be transmitted from the cluster to the base station.
Keywords – Base Station, Cluster, Cluster Head, Energyefficient, Hierarchical Routing Protocol, LEACH, Residual Energy, Wireless Sensor Network.
I.
For efficient utilization of sensor’s energy resources and maximizing the network lifetime many routing solutions for WSNs have been proposed [4][5] which consider the unique properties of the WSNs. Data aggregation in a hierarchical manner is widely used for prolonging network lifetime. Hierarchical mechanisms (especially clustering algorithms) are helpful to reduce data latency and increase network scalability [6]. Clustering techniques have emerged as a popular choice for achieving energy efficiency and scalable performance in large scale sensor networks. In this paper we have proposed an algorithm based on LEACH (LowEnergy Adaptive Clustering Hierarchy) which increases the overall energy utilization of the WSN.
INTRODUCTION
A wireless sensor network (WSN) is a group of sensor nodes which are deployed in a field to monitor physical conditions autonomously. WSNs can measure various physical conditions like sound, temperature, pressure, humidity, load, speed etc. After sensing the data sensor nodes pass this information to a base station or sink following a particular routing pattern. The number of sensor nodes in a WSN can vary from a few to hundreds or thousands in numbers depending on the application. A sensor node consists of many components, a microprocessor or a microcontroller to control the operation of node, a radio transceiver to transmit and receive information, an ADC converter to convert analog information to digital and vice versa and a power source. Batteries are normally used as power source in these sensors [16].
The rest of the paper is organized as follows: Section 2 describes the related works done in this area. We have explained the proposed protocol in Section 3. Section 4 compares the experimental results and evaluates the performance of LEACH and the proposed protocol. Finally, we conclude the paper in Section 5 with some scope of future work.
Unlike traditional wireless networks, sensor networks are characterized by power, computation, and memory constraints [3]. Since sensor nodes are generally deployed randomly in a remote location it is not feasible to provide power backup regularly. So the energy resource of sensor
978-1-4799-6758-2/14/$31.00 ©2014 IEEE
181
II.
RELATED WORK
The advantages of this approach are that no longdistance communication with the base station is required and distributed cluster formation can be done without knowing the exact location of any of the nodes in the network. In addition, no global communication is needed to set up the clusters and nothing is assumed about the current state of any other node during cluster formation. The goal is to achieve the global result of forming good clusters out of the nodes, purely via local decisions made autonomously by each node.
Our proposed algorithm is based on an energy-efficient hierarchical routing protocol LEACH [7]. The main aim of hierarchical routing is to efficiently maintain the energy consumption of sensor nodes by involving them in multihop communication within a particular cluster and by performing data aggregation and fusion in order to decrease the number of transmitted messages to the sink. LEACH [7] is one of the first hierarchical routing approaches for sensors networks. The idea proposed in LEACH has been an inspiration for many hierarchical routing protocols [16][20][21], although some protocols have been independently developed [11][15].
Some of the shortcomings of LEACH are that it does not consider the residual (remaining) energy of the CHs and it assumes uniform energy consumption for CHs which is not realistic. Let us consider a scenario [11] as described in Figure 1 in which most of the sensor nodes are grouped together around one or two cluster-heads. It is shown that cluster-heads A and B have more nodes close to them than the other cluster-heads. LEACH’s cluster formation algorithm will end up by assigning more cluster member nodes to both A and B. This could make cluster head nodes A and B quickly running out of energy [13] whereas the other two CHs will have significant amount of unused energy. This residual energy is wasted in LEACH because the CHs will be replaced in the next round and they will not get the chance of becoming CH again in the next 1/p rounds.
Low-Energy Adaptive Clustering Hierarchy (LEACH) [14] is one of the most popular hierarchical routing algorithms for sensor networks. It is a cluster-based protocol having the following features: randomized, adaptive, selfconfiguring cluster formation, localized control for data transfers, application-specific data processing, such as data aggregation. The idea of LEACH is to form clusters of the sensor nodes based on the received signal strength and use local cluster heads (CH) as routers to the sink i.e. all cluster members can transmit sensed data to the base station (BS) through CH only. This will save energy since the transmission to BS will only be done by cluster heads only instead of all the sensor nodes. Optimal number of cluster heads is estimated to be 5% of the total number of nodes. Operation of LEACH is based on rounds and each round consists of two phases - setup phase and steady state phase. In setup phase CHs and clusters are created. Some nodes independently elect themselves as CHs based on some probability P and their previous record as a CH. All nodes which were not CHs in previous 1/p rounds, generate a number from 0 to 1 and if it is less than a threshold T (n) then these nodes become CHs [7][9]. Threshold value is set by the given formula:
Fig. 1: A Sensor Network [13] T(n) = 0
: otherwise
(1)
III.
In formula (1) G is set of nodes that have not been selected as CHs in previous 1/p rounds, P is suggested percentage of CH, r is current round [10]. Elected CHs broadcast their status using CSMA/CA protocol. Non-CH nodes select their CHs by comparing the strength of received signals from multiple CHs. After creating clusters all CHs will create TDMA schedule for their associated members and broadcast it.
PROPOSED WORK
We have proposed an updated version of LEACH by considering the remaining energy of CH to use energy more efficiently and thereby increasing the network life time. Moreover like LEACH it is not realistic to consider that all sensor nodes in the network are homogeneous with respect to energy. Because it decreases the overall lifespan of the network as the cluster heads which directly communicate with the BS will drain out their energy earlier than the cluster member nodes. To make the network more energy efficient we have considered heterogeneous network with nodes having two different energy levels - normal node having less energy will be treated as the cluster members and advanced node possessing more energy will be elected
After that a steady state phase starts which is usually longer than the setup phase. In this phase nodes transmit data to their CHs during the allocated time slots otherwise they remain in sleep mode increasing battery lifetime. After receiving data from all the members CHs will aggregate the data and transmit to the BS [14].
182
as cluster heads. The CHs will aggregate the received data and transmit it either directly or indirectly through other CH to the BS. Whether the transmission will be direct or indirect that depends on the distance between the CH and the BS. Since energy consumption in wireless systems is directly proportional to the square of the distance [12], so much more energy is consumed when CHs directly transmit data to the BS. To reduce the energy consumption CHs will find the minimum distance to BS and transmit data through that path unlike LEACH where data is directly transmitted to the BS by the CHs irrespective of the distance.
4.
Clusters are formed depending on the signal strength a normal node receives from different CHs.
5.
The normal nodes send a join message to the corresponding cluster heads that in turn create TDMA schedule for data transmission and broadcast it to the members.
Steady-state Phase:
As we have already mentioned that another shortcoming of LEACH is that some of the CHs are overloaded whereas some are under-loaded. So it is not a wise decision to replace all the CHs together irrespective of their remaining energy level; because once it has become a CH, it will not get another chance of being a CH in the next 1/p rounds. So the under-utilized CH must get another chance for continuing as a CH in the next round also. This method will not only use the remaining CH energy but also decrease the overhead associated with a new CH formation in each round. Since if a new CH is selected then whole process of cluster formation will take place.
1.
Like LEACH all the cluster members will send data to their corresponding cluster-heads in their allotted time slot.
2.
Cluster Heads aggregate the received data and transmit it directly or indirectly through other CH to the BS.
3.
Once all the Cluster Heads finish the control returns to steady phase again IV.
PERFORMANCE EVALUATION
We simulated LEACH, MOD-LEACH and the proposed LEACH-based protocol to make efficient analysis. In MODLEACH [17] dual transmitting power levels to amplify signals according to nature of transmission are introduced. We compare our protocol with it because it also uses residual energy for cluster head replacement scheme.Simulation parameters are shown in table 1. This simulation is implemented by using MATLAB (version 7.10). It shows that the proposed protocol performs better considering the metrics of network lifetime and throughput.
Initially the CHs will be selected from the set of advanced nodes following the same process of LEACH. The CH selection formula depicted in equation (1) will be used here also. But during reclustering a threshold value will be used to decide whether the CH will be replaced or not. The threshold defines the minimum energy level which must be possessed by a CH to communicate with the BS directly. Here for simplicity we have considered static threshold which is equal to be half of the initial energy of the normal nodes. But it will be more realistic to consider dynamic threshold which will be adjusted automatically depending on the average residual energy of the network.
TABLE I Simulation Parameters Parameter
Value
If the residual energy in the CH is greater than the threshold then the CH will not be changed. If cluster head has less energy than the threshold then it will be replaced according to the LEACH protocol.
Network Size
100m * 100m
Number of nodes
100
Pseudo code for the proposed protocol –
Packet Size
4000 bits
Set-up Phase:
Initial Energy of normal nodes
0.5J
Initial Energy of advanced nodes
1.0J
Number of rounds
3000
Transmitter Electronics (ETX)
50nJ/bit
Receiver Electronics (ERX)
50nJ/bit
Data Aggregation Energy
5nJ/bit
1.
2.
3.
In the first round all the advanced nodes generate a random number between 0 and 1 and if it is less than the threshold T(n) (equation 1) then the node is elected as a cluster head for that round. Goto step 3. In all other rounds the remaining energy of cluster heads are checked. If it is greater than a predefined threshold then it will continue as the cluster head in the next round also. So go to steady-state otherwise select new cluster head as described in step 1. CHs broadcast hello message.
183
Result The simulated results depicted in Figure 2 and Figure 3 compare network life time of LEACH, MODLEACH and proposed protocol by showing number of dead and alive nodes respectively. In the proposed method the first node dies earlier than LEACH because the same CH is used in successive round if it possesses sufficient amount of energy. So a few nodes may drain out their energy completely and die earlier than the nodes in LEACH. In MODLEACH the nodes started dying later due to the efficient use of dual amplification energy for transmission of data. But in spite of that the proposed method increases the lifetime of WSN by 7.16% and 51.46% compare to LEACH and MODLEACH respectively as shown in Table II.
Besides network life time the other parameter which we have considered is the throughput of the routing protocols. The more packets are received by the base station the more efficient is the routing protocol. From the simulated results shown in Figure 4 and Figure 5 we can conclude that the proposed routing technique achieves more throughput than LEACH. This is basically due to the proper use of residual energy by the CHs in the proposed protocol. As in this protocol from second round onwards the set-up phase will take less time compare to that of LEACH. So the time duration of steady-phase will increase in each round which in turn increase the amount of data transferred to the CH and BS.
Fig2: Dead nodes in LEACH, MODLEACH and proposed protocol
Fig4: Packets to CH in LEACH, MODLEACH and proposed protocol
Fig3: Alive nodes in LEACH, MODLEACH and proposed protocol
Fig5: Packets to BS in LEACH, MODLEACH and proposed protocol
184
From Figure 4 we can see that in MODLEACH more packets are transmitted to CH compare to the proposed protocol. This is because of dual transmitting power levels within network. Use of different amplification energies for transmissions decreases packet drop ratio resulting in higher throughput. Packets transferred to BS are more in case of proposed protocol because MODLEACH uses soft and hard threshold values to restrict transmission.
Whereas single hop LEACH is not suitable for such network due to the limit in effective communication range of the sensor nodes [15]. The future work includes finding an optimal value for the dynamic threshold which can be used in the proposed method for energy efficient replacement of cluster head. The concept of mobility can be further incorporated to our algorithm. We believe that the propose method can offer significant improvement on the performance of sensor networks if further research is carried out on the above mentioned issues.
From the simulation results shown above, an improvement table is drawn below which shows lifetime improvement on the basis of first node dead and all nodes dead.
REFERENCES
Table II shows the rounds when nodes start dying and all the nodes are dead in LEACH, MODLEACH and proposed protocol.
[1]
TABLE II Comparison of the protocols Protocol LEACH MODLEACH Proposed Protocol
[2]
Rounds when Rounds when all nodes start dying nodes are dead 828 2178 1053 1541 667 2334
[3]
[4]
As we have already mentioned that in the updated algorithm nodes started to die earlier due to the nonreplacement of all the CHs after the first round. But the strategy works well since the overall lifetime of WSN is increased by 7.16% than LEACH which can be explained mathematically as below:
[5]
[6]
((Last Round of Proposed Protocol – Last Round of LEACH)/Last Round of LEACH) * 100
[7]
= ((2334-2178)/2178)*100 = 7.16% Similarly, 51.46% improvement over MODLEACH in overall lifetime of WSN is obtained by our proposed algorithm. V.
[8]
[9]
CONCLUSION AND FUTURE WORK
The paper presents an energy efficient clustering algorithm based on LEACH for wireless sensor network. We have simulated the proposed algorithm on MATLAB and compare the performance of LEACH, MODLEACH and the proposed protocol. Simulation shows that the lifetime and throughput of our proposed method are more than that of LEACH and MODLEACH. This improvement is achieved by the heterogeneity of nodes and efficient use of remaining energy of the cluster heads which remains unused in LEACH. The cluster heads communicate with the base station directly or indirectly using multihop hierarchical routing through other CHs which makes our algorithm suitable for large-scale wireless sensor networks.
[10]
[11]
[12]
185
Shio Kumar Singh, M P Singh, D K Singh, “A Survey of Energy-Efficient Hierarchical Cluster-Based Routing in Wireless Sensor Networks”, Int. J. of Advanced Networking and Applications Volume: 02, Issue: 02, 2010, pp. 570-580. B. Krishnamachari, D. Estrin and S. Wicker, “The impact of data aggregation in wireless sensor networks”, Proceedings of IEEE International Conference on Distributed Computing Systems Workshops, July 2002, pp. 575–578. Guihai Chen, Chengfa Li, Mao Ye, Jie Wu , “An unequal cluster-based routing protocol in wireless sensor networks ”, Wireless Netw, Springer , 2009, pp. 193–207 . J. N. Al-Karaki and A. E. Kamal, “Routing Techniques in Wireless Sensor Networks: A Survey,” IEEE Journal of Wireless Communications, Vol. 11, No. 6, 2004, pp. 6-28. A. Martirosyan, A. Boukerche and R. W. N. Pazzi, “A Taxonomy of Cluster-Based Routing Protocols for Wireless Sensor Networks,” The International Symposium on Parallel Architectures, Algorithms, and Networks Sydney, May 2008, pp. 247-253. G. J. Pottie and W. J. Kaiser, “Wireless Integrated Network Sensors,” Communications of the ACM, Vol. 43, No. 5, 2000, pp. 51-58. W. Heinzelman, A. Chandrakasan and H. Balakrishnan, "Energy-Efficient Communication Protocol for Wireless Microsensor Networks," Proceedings of the 33rd Hawaii International Conference on System Sciences (HICSS '00), January 2000. Subhadra Shaw (Bose), “Energy-Efficient Routing Protocol in Wireless Sensor Network”, International Journal of Scientific & Engineering Research Volume 2, Issue 12, ISSN 22295518, December-2011. M. Aslam, M. B. Rasheed, T. Shah, A. Rahim, Z. A. Khan, U. Qasim, M. W. Qasim, A. Hassan, A. Khan, N. Javaid , “Energy optimization and Performance Analysis of Cluster Based Routing Protocols Extended from LEACH for WSNs”, X. H. Wu, S. Wang, “Performance comparison of LEACH and LEACH-C protocols by NS2,” Proceedings of 9th International Symposium on Distributed Computing and Applications to Business, Engineering and Science. Hong Kong, China, pp. 254-258, 2010. Lan Tien Nguyen, Xavier Defago, Razvan Beuran, Yoichi Shinoda. An Energy Efficient Routing Scheme for Mobile Wireless Sensor Networks; 568-569. Lalit Kumar Saraswat, Dr. Sachin Kumar, “Resource Biased Routing (RBR) Algorithm for Energy Optimization in Wireless Sensor Networks”, International Journal of Engineering and Innovative Technology (IJEIT) Volume 2, Issue 3, September 2012, pp. 180-184.
[13] Lan Tien Nguyen, Xavier Defago, Razvan Beuran, Yoichi
Sensor Networks: A Survey”, IEEE COMMUNICATIONS SURVEYS & TUTORIALS, VOL. 15, NO. 2, SECOND QUARTER 2013, pp. 551-591. [16] Wairagu G. Richard, “Extending LEACH routing algorithm for Wireless Sensor Network,” Data Communications Engineering, Makerere University, 2009. [17] D. Mahmood, N. Jayaid, S. Mahmood, S. Qureshi, A. M. Memon, T. Zaman, “MODLEACH: A Variant of LEACH for WSN”, 26th IEEE Canadian Conference on Electrical and Computer Engineering (CCECE-2013), 2013.
Shinoda, “An Energy Efficient Routing Scheme for Mobile Wireless Sensor Networks”, IEEE ISWCS 2008, pp. 568-572. [14] MEENAKSHI SHARMA AND ANIL KUMAR SHAW, “Transmission Time and Throughput analysis of EEE LEACH, LEACH and Direct Transmission Protocol: A Simulation Based Approach”, Advanced Computing: An International Journal (ACIJ), Vol.3, No.5, September 2012. [15] Nikolaos A. Pantazis, Stefanos A. Nikolidakis and Dimitrios D. Vergados, “Energy-Efficient Routing Protocols in Wireless
186
