Weighted Clustering using Comprehensive Learning Particle Swarm Optimization for Mobile Ad Hoc Networks

A mobile Ad-hoc network consists of dynamic nodes that can move freely. These nodes communicate with each other without a base station. In this paper, we propose a Comprehensive Learning Particle Swarm Optimization (CLPSO) based clustering algorithm for mobile ad hoc networks. It has the ability to find the optimal or near-optimal number of clusters to efficiently manage the resources of the network. The cluster-heads do the job of routing network packets within the cluster or to the nodes of other clusters. The proposed CLPSO based clustering algorithm takes into consideration the transmission power, ideal degree, mobility of the nodes and battery power consumption of the mobile nodes. It is a weighted clustering algorithm that assigns a weight to each of these parameters of the network. Each particle of the swarm contains information about the cluster-heads and the members of each cluster. It uses the evolutionary capability to optimize the number of clusters. We compare the simulation results with two other well-known clustering algorithms. The results show that the proposed technique is effective and works better than the other two approaches.