Low Energy Fuzzy based Unequal Clustering Multihop Architecture for Wireless Sensor Networks

Abstract

This paper presents a Low Energy Fuzzy based Unequal Clustering Multihop Architecture (LEFUCMA) for wireless sensor networks consisting of several nodes that send sensed data to a Master Station (MS). LEFUCMA encompasses neighbor finding, cluster head selection, clustering and routing protocols. The neighbor finding protocol organizes the network into a sectored-layers structure. The cluster head selection uses fuzzy logic with residual energy, number of neighboring nodes, packet reception rate and distance of node from MS as fuzzy descriptors for cluster head selection. For even distribution of traffic, LEFUCMA uses fuzzy logic with node density and distance of area from MS as fuzzy descriptors to decide number of cluster heads in a given area. To avoid hot spots problem, LEFUCMA uses an unequal clustering mechanism with clusters closer to MS having smaller sizes than those farther from MS. Finally, inter-cluster routing protocol decides the next hop cluster head considering its residual energy, distance from MS and from current cluster head which represents energy required for communication, number of cluster members which represents intra-cluster traffic and number of descendant nodes which represents inter-cluster traffic. A comparative analysis of LEFUCMA; Unequal Hybrid, Energy Efficient and Distributed Clustering (Ever et al. in Proceedings of international conference on sensor networks, pp 185–193, 2012; Energy Aware Distributed Unequal Clustering (Yu et al. in Hindawi Int J Distrib Sens Netw, Article ID 202145:1–8, 2011); Constructing Optimal Clustering Architecture (Li et al. in J Comput Commun 36(3):256–268, 2013; and Energy Aware Unequal Clustering using Fuzzy logic (EAUCF) (Bagci and Yazici in J Appl Soft Comput 13(4):1741–1749, 2013) shows that LEFUCMA is 32–42% more energy efficient compared to EAUCF. Throughput of LEFUCMA is 46% more and network lifetime is 60–75% more compared to EAUCF.