Equitable Leach-E Protocol for HeterogeneousWireless Sensor Networks

A. Beni hssane, My. L. Hasnaoui, M. Saadi, S. Benkirane, and M. Laghdir

Abstract. A clustering algorithm is a key technique used to increase the scalabilityand lifetime of the Wireless Sensor Networks (WSNs). In this paper, we proposeand evaluate a distributed energy-efficient clustering algorithm for heterogeneousWSNs, which is called Equitable LEACH-E (ELE). This protocol is an improve-ment of LEACH-E. In ELE, the cluster-heads are elected by using probabilitiesbased on the ratio between residual energy of each node and the remaining energyof the network. Moreover, it uses a 2-level hierarchy by selecting a cluster headfor data transmission. Simulations show that the proposed algorithm increases thelifetime of the whole network and performs better than LEACH and LEACH-E.

Keywords: Wireless Sensor Networks; Clustering Algorithm; HeterogeneousEnvironment; Energy-Efficient.

1 Introduction

Owing to the advances of Micro-Electro-Mechanical Systems (MEMS) and wirelesscommunication, Wireless Sensor Networks (WSNs) have become an indispensabletool to carry out many applications impossible for other types of networks [1, 2]. AWSN is composed of a large number of sensor nodes that are deployed in ad hocmanner. Clustering technique is used to increase the lifetime of WSNs [3, 4]. In fact,only some nodes are required to transmit data over a long distance and the rest willneed to complete short distance transmission only.

A. Beni hssane · My. L. Hasnaoui · M. Saadi · S. Benkirane · M. LaghdirChouab Doukkali University, Faculty of Sciences Department of Mathematics andComputer Science, MATIC laboratory, El Jadida, Moroccoe-mail: abenihssane@yahoo.fr, mlhnet2002@yahoo.ca,saadi mo@yahoo.fr, sabenk1@hotmail.com, laghdirm@yahoo.fr

M. Essaaidi et al. (Eds.): Intelligent Distributed Computing IV, SCI 315, pp. 171–176.springerlink.com c© Springer-Verlag Berlin Heidelberg 2010

172 A. Beni hssane et al.

In [5], it is proposed to elect the cluster-heads according to the energy left in eachnode. In [6], this clustering protocol is called LEACH-E. Based on LEACH-E pro-tocol, we develop and validate a newer Equitable LEACH-E algorithm called ELE.This protocol is proposed to increase the whole network lifetime on a heterogeneousnetwork with a BS located far away from the sensor area. ELE introduces the 2-levelhierarchy concept based on the maximum of the ratio between residual energy andthe distance to the BS of each cluster head. This permits a better distribution of theenergy load through the sensors in the network.

The remainder of this paper is organized as follows. Section 2 presents the relatedwork. Section 3 exhibits the details and analyzes the properties of ELE. Section 4evaluates the performance of ELE by simulations and compares it with LEACH andLEACH-E. Finally, Section 5 gives concluding remarks.

2 Related Work

The routing protocols for WSNs can be categorized as follow: the clusteringalgorithms applied in homogeneous networks are called homogeneous schemes,where all nodes have the same initial energy, such as LEACH [4]; and the cluster-ing algorithms applied in heterogeneous networks are referred to as heterogeneousclustering schemes, where all the nodes of the sensor network are equipped with dif-ferent amount of energy, such as SEP[7], EECS[8], DEEC[6], and [9]. The EECSprotocol elects the cluster-heads with more residual energy through local radio com-munication. The DEEC protocol is a distributed energy-efficient clustering schemefor heterogeneous wireless sensor networks, in which the cluster-heads are electedby a probability based on the ratio between residual energy of each node and theaverage energy of the network. In DEEC protocol, the BS is located in the center ofthe sensing area and uses 1-level hierarchy concept. In [5], a protocol is proposedto elect the cluster-heads according to the energy left in each node. This protocol iscalled LEACH-E in [6]. The drawbacks of LEACH-E are that it requires the assis-tance of routing protocol, which should allow each node to know the total energy ofnetwork and it utilizes direct transmission from a cluster heads to the BS.

Our work is inspired by the previous approaches, but it differs by introducing toLEACH-E an adapted formulas to estimate the network lifetime, thus avoiding theneed of assistance by routing protocol. Since we assume that the BS is far awayfrom the sensing area, we are using 2-level hierarchy concept for transmitting datato the BS.

3 Equitable LEACH-E (ELE)

Our ELE uses the same clustering algorithm, the same strategy in Clusters Headselection, Clusters formation, and Schedule Creation (TDMA)as LEACH-E but dif-fers in Data transmission. ELE algorithm can be summarized as in figure Fig 1.

Since the BS is located far away from the network, the total network energyconsumption in each transmission to the BS will be very important. To optimize

Equitable Leach-E Protocol for Heterogeneous Wireless Sensor Networks 173

the energy dissipated in the network, our ELE introduces 2-level hierarchy conceptwhich allows a better use of the energy consumed in the network. In ELE, the prob-ability threshold, that each node si uses to determine whether itself to become acluster-head in each round, is given as follow [6]:

T (si) =

{ pi

1−pi(rmod 1pi

), i f si ε G

0 , otherwise(1)

where G is the set of nodes that are eligible to be cluster heads at round r. In eachround r, when node si finds it is eligible to be a cluster head, it will choose a randomnumber between 0 and 1. If the number is less than threshold T (si), the node si

becomes a cluster head during the current round. Also, pi is defined as follow [5]:

pi(r) = min

{Ei(r)

Etotal(r)k , 1

}(2)

where Ei(r) is the current energy of node i, k is the desired number of cluster, andEtotal(r) is an estimate of the remaining energy of the network per round r:

Etotal(r) = Einitial

(1− r

R

)(3)

where R denotes the total rounds of the network lifetime. The value of R is:

R =Einitial

ERound(4)

where ERound, denotes the total energy dissipated in the network during a round r, isgiven by:

ERound = L[2NEelec + NEDA +(k−1)εmpd4toMax RatCH + Nε f sd

2toCH + Eεmpd4

toBS](5)

where k is the number of clusters, EDA is the data aggregation cost expended in thecluster-heads, dtoBS is the average distance between the cluster-head and the BS,dtoMax RatCH is the average distance between the cluster-heads and the Max RatCH,which is the CH that has the maximum ratio between residual energy and the dis-tance to the BS of each CHs, dtoCH is the average distance between the clustermembers and the cluster-head. Assuming that the nodes are uniformly distributedand by using the result in [5, 10], we can get the equations as follow:

dtoCH =M√2kΠ

(6)

dtoMax RatCH =1

M2

∫ ∫ √(xi + x j)2 +(yi + y j)2dxdy ≈ M

2(7)

dtoBS =√

2ΠM2

(8)

174 A. Beni hssane et al.

k =√

ε f s

εmp

√N

2ΠM

d2toMax RatCH

(9)

Substituting equations ( 9, 8, 7, 6, 5, 4, 3, and 2) into equation ( 1), we obtainthe probability threshold. Based on the information coordinates and the residualenergy included on the message broadcasted, the CHs elected can select one of themwhich has the Maximum of the Ratio between residual energy and the distance tothe BS. We called this intermediate Cluster Head as Max RatCH. The ratio can beseen as trade off between residual energy and nearness to the BS. The Max RatCHcollects all data coming from all CHs, compress it into a single signal and send itdirectly to the BS. Each non cluster heads(NCH) sends its data during their allocatedtransmission time (TDMA) to the respective cluster head. The CH node must keepits receiver on in order to receive all the data from the nodes in the cluster. When allthe data is received, the cluster head node performs signal processing functions tocompress the data into a single signal. When this phase is completed, each clusterhead can send the aggregated data to the Max RatCH.

Fig. 1 Algorithm

4 Simulation Results

We consider a WSN with N = 100 nodes randomly distributed in a 100m× 100msensing area. We assume the BS is far away from the sensing region and placed atlocation (x = 50,y = 175). The radio parameters used in our simulations are shownin Table 1. We assume that all nodes know their location coordinates. We use inthis study a similar energy model as proposed in[5]. We will consider the followingscenarios and examine several performance measures.

First, we observe the performance of LEACH, LEACH-E, and ELE under twokinds of 2-level heterogeneous networks. Fig. 2 shows the results of the case withm = 0.1 and a = 5. It is obvious that the stable time of ELE is prolonged comparedto that of LEACH and LEACH-E.

Second, we run simulation for our proposed protocol ELE to compute the roundof the first node dies when m is varying and compare the results of to LEACH

Equitable Leach-E Protocol for Heterogeneous Wireless Sensor Networks 175

Fig. 2 Number of nodes alive over time(a = 5, m = 0.1) Fig. 3 Round first node dies when m is

varying

Fig. 4 Number of message received at theBS over time (a = 5, m = 0.1)

Fig. 5 Number of message received at theBS over energy spent(a = 5, m = 0.1)

and LEACH-E protocols. We increase the fraction m of the advanced nodes from0.5 to 5, Fig. 3 shows the number of round when the first node dies. LEACH-Eperforms well and achieves the stability period longer by about 100% than LEACH.ELE outperforms LEACH-E protocol. In fact, when m is varying, ELE obtains 41%number of round than LEACH-E.

Third, Fig. 4 shows that the number of delivered messages to the BS by ELEprotocol are greater than the others ones; this means that ELE is a more efficientprotocol.

Fourth, Fig. 5 shows that ELE is a more efficient of energy consumption.

5 Conclusion

ELE is proposed as an energy-aware adaptive clustering protocol used in hetero-geneous WSNs. To control the energy expenditure of nodes by means of adaptive

176 A. Beni hssane et al.

approach, ELE uses the average energy of the network as the reference energy. Thus,ELE does not require any global knowledge of energy at every election round.Moreover, ELE uses the 2-level hierarchy concept which offers a better use andoptimization of the energy dissipated in the network. Finally, the introduced modi-fications enlarge and outperform better the performances of our ELE protocol thanthe LEACH and LEACH-E.

References

1. Estrin, D., Girod, L., Pottie, G., Srivastava, M.: Instrumenting the world with wirelesssensor networks. In: Proceedings of the International Conference on Acoustics, Speechand Signal Processing (ICASSP 2001), Salt Lake City, Utah, USA, May 2001, vol. 4, pp.2033–2036 (2001)

2. Akyildiz, F., Su, W., Sankarasubramaniam, Y., Cayirici, E.: A survey on sensor networks.IEEE communications magazine 40(8), 102–114 (2002)

3. Mhatre, V., Rosenberg, C., Kofman, D., Mazumdar, R., Shroff, N.: Design of surveillancesensor grids with a lifetime constraint. In: Karl, H., Wolisz, A., Willig, A. (eds.) EWSN2004. LNCS, vol. 2920, pp. 263–275. Springer, Heidelberg (2004)

4. Heinzelman, W.R., Chandrakasan, A.P., Balakrishnan, H.: Energyefficient communica-tion protocol for wireless microsensor networks. In: Proceedings of the 33rd HawaiiInternational Conference on System Sciences (HICSS-33) (January 2000)

5. Heinzelman, W.R., Chandrakasan, A.P., Balakrishnan, H.: An application-specific pro-tocol architecture for wireless microsensor networks. IEEE Transactions on WirelessCommunications 1(4), 660–670 (2002)

6. Qing, L., Zhu, Q., Wang, M.: Design of a distributed energy-efficient clustering al-gorithm for heterogeneous wireless sensor networks. Computer Communications 29,2230–2237 (2006)

7. Smaragdakis, G., Matta, I., Bestavros, A.: SEP: A Stable Election Protocol for clusteredheterogeneous wireless sensor networks. In: Second International Workshop on Sensorand Actor Network Protocols and Applications, SANPA 2004 (2004)

8. Ye, M., Li, C., Chen, G., Wu, J.: EECS: an energy efficient cluster scheme in wirelesssensor networks. In: IEEE International Workshop on Strategies for Energy Efficiencyin Ad Hoc and Sensor Networks (IEEE IWSEEASN 2005), Phoenix, Arizona, April 7-9(2005)

9. Liu, M., Cao, J., Chen, G., Wang, X.: An Energy-Aware Routing Protocol in WirelessSensor Networks. Sensors 9(1), 445–462 (2009)

10. Bandyopadhyay, S., Coyle, E.J.: An energy efficient hierarchical clustering algorithm forwireless sensor networks. In: Proceeding of INFOCOM 2003 (April 2003)