A Multi-metric-Based Algorithm for Cluster Head Selection in Multi-hop Ad Hoc Network

Clustering technique finds wide applicability in wireless networks. Cluster structures enhance resource reuse as well as increases system capacity. Clustering is a desirable task in MANET if the size of multi-hop wireless network becomes too large. In the existing work, number of cluster formation depends on the radius of cluster chosen in a scenario. But here, we propose a new approach in cluster head selection within a cluster. Here, we consider three parameters, i.e. number of node neighbours, node lifetime and stability of nodes which in turn are used for selection of an appropriate cluster head. We simulated our algorithm and measured performance by considering energy consumption, throughput and control overhead as performance metrics. Our approach has been found outperforming to the existing ones.