A Game Theoretic Approach for Energy-Efficient Clustering in Wireless Sensor Networks

Selection of clusterheads using energy efficient clustering algorithms in a wireless sensor network (WSN) is very crucial as it affects the lifetime and performance of the network. As clusterheads and cluster members (i.e., non-clusterhead nodes) have a different energy consumption rates, it is necessary that all nodes resort to some rational scheme such that the connectivity and proper functioning of the network is not compromised. In this paper, we propose a Cost and Payment-based clustering Algorithm (CoPA) for achieving energy efficiency in wireless sensor networks under a game theoretical framework. The analysis is based on a non-cooperative, repeated general-sum game, where each node behaves selfishly in order to maximize its lifespan (payoff). We demonstrate that the correlated equilibrium is a practical solution for clusterhead selection, which provides better performance than the Nash Equilibria. Correlated equilibrium provides a balance between the fully cooperative solution and the fully non-cooperative solution in terms of implementation overhead. CoPA produces a balanced distribution of responsibilities and energy consumption between the sensor nodes as well as maximizes the minimum payoff for every node. Results show that CoPA achieves better performance in terms of network lifetime and throughput compared to other popular clustering techniques.