Mobility-sensitive clustering algorithm to balance power consumption for mobile ad hoc networks

Abstract

Energy efficiency is very important in mobile ad hoc networks (MANETs). Most clustering algorithms save energy by limiting the maximum size of the cluster, by reducing control overhead for maintaining clusters, or by reducing power consumption of intra-cluster communication. These algorithms usually update the intra-cluster routing information and maintain clusters by periodically sending out beacons. This paper proposes to save energy not only by reducing the unnecessary flooding used to update the intra-cluster routing information, but also by balancing power consumption among nodes and clusters throughout the network. We first propose a method to assess the mobility rate of the nodes by sensing the difference in node ID records in a CH between two successive beacons. The beacons will then be adaptively sent out to update the intra-cluster routing information according to these assessed mobility rates, so that the unnecessary flooding is significantly reduced, thereby, resulting in higher energy efficiency. Furthermore, the cluster sizes are adaptively maintained between the maximum and the minimum sizes according to the mobility rates to better balance power consumption among the clusters. Preliminary simulation results show that the proposed algorithm indeed sends out beacon messages adaptively and balances power consumption among clusters.