点击次数：

论文类型： 期刊论文

发表时间： 2010-10-01

发表刊物： ICIC Express Letters

收录刊物： EI、Scopus

卷号： 4

期号： 5

页面范围： 1555-1560

ISSN号： 1881803X

摘要： A novel collaborative tracking algorithm based on distributed particle filter is proposed for tracking mobile target in wireless sensor networks. Firstly, we present a dynamic cluster architecture, where the schemes to select the cluster heads and the cluster member nodes are designed. Secondly, a distributed particle filter algorithm is proposed to collaboratively realize computation among cluster members, where the unscented particle filter algorithm is running simultaneously in all the cluster members so as to speed up the computation, and the Gaussian mixture model is adopted to approximate the posterior distribution of the weighted particles so as to reduce the communication payload among the cluster heads. Simulation results validate the algorithm and demonstrate that its performance is close to the posterior Cramer-Rao lower bound. ICIC International ? 2010 ISSN.