Differential game based centralized clustering algorithm for wireless sensor networks

Clustering is an effective approach for managing nodes in wireless sensor networks (WSN). In this paper, we consider selfish avoidance clustering in WSN. A differential game model is proposed to stimulate forwarding. Based on the payoff value of the differential game, called forwarding contribution (FC) in this paper, a centralized clustering algorithm is proposed. Mathematical proof shows that our algorithm can ensure connection between cluster heads (CHs), and also shows that FC value can reasonably reflect energy consumption in terms of competition while considering forwarding efficiency. Simulation on the lifetime of network is performed, and the results illustrate that our algorithm is selfish avoidance while the clustering approaches according to residual energy are seriously influenced by selfishness.