A noble target tracking algorithm based on dynamic clustering for wireless sensor network

Abstract

This paper proposes a dynamic clustering algorithm for target tracking to balance energy consumption in a local monitoring region of wireless sensor networks. In order to balance the local energy consumption of a dynamic clustering owing to excessive use of a individual node, one node is preferentially chosen as a tasking node which is marked by a high value of the decision function and the node with the highest value is chosen as the cluster head, and others with a low value are in reserve. In addition to the joint detection probability (JDP) for tasking sensors, particle filter algorithm is used for target tracking in every chosen node. Simulation results show that, the algorithm can achieve the required tracking accuracy, and solve the problem that nodes are reused so as to balance the consumption of energy and prolong the lifetime of wireless sensor networks.