Clusterhead election for mobile ad hoc wireless network

Abstract

Selecting an appropriate clusterhead can save power for the whole ad hoc wireless network. Generally, clusterhead election for mobile ad hoc network is based on the distance to the centroid of a cluster, and the closest one is elected as the clusterhead; or pick a node with the maximum battery capacity as the clusterhead. In this paper, we present a clusterhead election scheme using fuzzy logic system (FLS) for mobile ad hoc wireless networks. Three descriptors are used: distance of a node to the cluster centroid, its remaining battery capacity, and its degree of mobility. The linguistic knowledge of clusterhead election based on these three descriptors is obtained from a group of network experts. 27 FLS rules are set up based on the linguistic knowledge. The output of the FLS provides a clusterhead possibility, and node with the highest possibility is elected as the clusterhead.