A collaborative tracking algorithm for communicating target in Wireless Multimedia Sensor Networks

In this paper, we address the problem of target tracking in Wireless Multimedia Sensor Networks. Target tracking is usually defined as a two stages process: 1) detecting the presence of the target and 2) locating it. We propose a cluster-based and collaborative tracking algorithm for a signal emitting target with the objective of finding the best trade-off between the energy consumption and the tracking precision. In this algorithm, each cluster component is in charge of specific tasks. More powerful sensors handle the high cost energy tasks and assume inter and intra-cluster collaboration while constraint sensors handle low-cost energy tasks and assume only intracluster communication. A probabilistic node selection method is implemented to select the best sensors which participate to the tracking process. A deployment strategy for both sensors is also proposed. Simulation results are presented to evaluate the efficiency of the proposed algorithm. They demonstrate a significant target tracking accuracy improvement and energy consumption reduction comparing to existing algorithms.