Energy Efficient Clustering Techniques in Heterogeneous Wireless Sensor Networks

Abstract

Wireless sensor network is consisting of power-limited sensors that sense the surrounding environment and transmit data to another sensor or to the particular gateway. Thus, a considerable amount of energy is dissipated which impact the overall performance and network lifetime. In order to mitigate the failure-based energy, routing and clustering mechanisms should be energy aware. Accordingly, and in the literature, many hierarchical clustering techniques are proposed in term to extend the network alive time such as LEACH and SEP. The role of the elected cluster head is the main objective of these techniques and appear the differences between them. In this paper, two energy efficient clustering techniques are evaluated in 100 uniformly distributed wireless sensor networks in term of number of alive nodes, residual energy, and stability period. The election of cluster head is achieved in heterogeneous conditions where some of the nodes are powered higher than others. The results show how the heterogeneity of wireless sensor network in term of energy remaining at nodes influences the overall performance and increase the stability region before first node is dying. SEP is outperformed LEACH in term of long stability zone by 4.2% when 20% of the nodes (m=0.2) are energized more energy by 100% (α=1). Moreover, when m=0.2 and α=.3, SEP has better stable region than LEACH by 26%. In general, the election of cluster head mechanism in LEACH clustering leads to reduce the stable region and the node die faster comparing to SEP. In which, the cluster head election is totally based on the remaining energy.