Optimal number of cluster heads for random topology WSNs using the stable election protocol

The primary issue in Wireless Sensor Networks (WSNs) is to minimize energy consumption to extend network lifetime. Generally, sensor nodes are distributed randomly. Clustering technique helps to prolong the lifetime of a sensor network. It can also increase network scalability. In these networks, a few nodes are elected as Cluster Heads (CHs) whose main goal is to aggregate data from member nodes and transmit it to the Base Stations (BS) in order to minimize transmission and thus, energy consumption. Stable Election Protocol (SEP) introduces heterogeneity in WSNs. With SEP, nodes are distinguished based on their weighted election probabilities for becoming cluster heads, according to their residual energy. In this paper, we study the SEP protocol and we show by means of simulation and analytical formulas how to obtain the optimal number of elected cluster heads for random network topologies and various sizes.