A Reaction-Diffusion and Gür Game Based Routing Algorithm for Wireless Sensor Networks

In this paper, we propose an energy-efficient, cluster-based routing algorithm to address the issue of energy constraints in wireless sensor networks. There are two components in the proposed model, the first supports the development of clusters and the second helps decide which of the sensors will sleep. Together they improve the lifetime of the clusters. Biologically inspired activator-inhibitor mechanism is employed to form clusters and select cluster heads based on the activator concentration where each sensor is associated with a pair of activator and inhibitor concentration values. In each cluster, a Gür game is applied to determine the set of active sensor nodes while inactive sensor nodes turn to sleep mode for conserving energy. The activator–inhibitor system is known to provide the mechanism for autonomous biological pattern formation, such as spots on mammals’ coats, through interactions between molecules and their diffusion rates. The Gür game is a self-organized artificial game associating voters in the game with finite state automata and a moderator with a reward function. Typically in wireless sensor networks, the base station is considered as the moderator and sensor nodes as voters in the Gür game. To further maximize the lifetime of the network, in our proposed routing algorithm, each cluster is then associated with a Gür game to determine the number of active sensor nodes where the cluster head is regarded as the moderator and the cluster members as voters. Finally, we present preliminary results on the comparison between the proposed routing algorithm and LEACH, a well-known distributed clustering protocol used in wireless sensor networks that shows our method works better than LEACH.