Battery Power Clustering Using Ant Colony Optimization

Abstract

Wireless ad-hoc networks have limited resources such as power, memory, therefore, developing clustering algorithms that consider battery power for wireless node is indeed necessary in order to prolong network life time and decrease the probability of network partitioning. This work proposes a Battery Power Clustering (BPC) technique based on using ant colony optimization (ACO) method, in order to select Cluster Head (CH) in mobile ad hoc networks (MANETs). BPC technique considers battery power of each candidate CH Mobile Node (MN) and the size of clusters (number of neighboring MNs). Moreover, the influence of the battery power and the cluster size on CH election is optimized using ACO algorithm, BPC uses the number of transmitted signals – packets as a gauge for the battery power in candidate CHs. BPC is compared with the two hop Linked-Cluster Algorithm (LCA) and the Weighted Clustering Algorithm (WCA), comparisons employ Ad-hoc On-demand Distance Vector routing (AODV), Wireless Routing Protocol (WRP), and Location-Aided Routing (LAR1), such that the evaluation is conducted for different mobility speeds. Simulation results reveal the robustness of the proposed BPC mechanism in terms of average delay and average packets drop ratio.