Bounded clustering with low node-clusterhead separation in wireless sensor networks

Clustering is an important technique for distributed coordination in sensor networks. Clusters of bounded size are of special interest for their effectiveness in reducing energy consumption by limiting the neighborhood of a node. Further energy saving can be achieved by reducing the average node to clusterhead separation as it allows nodes to transmit at lower power. In this paper, we propose a bounded clustering algorithm which reduces the average node-clusterhead separation for energy-efficient communication. Extensive simulation results show that the algorithm generates bounded clusters where average node-clusterhead separation is low. In addition, for about 80% of the nodes, the actual separation is less than the average.