By analyzing the inherent relationship among the target tracking monitoring precision
node energy consumption and cluster member wake / sleep in WSNs
partially observable Markov decision process(POMDP) theory is imported
a dynamic cluster member scheduling model based on node energy consumption weighted reward rate and target tracking accuracy is proposed to address the problem of limited energy
monitoring data redundancy from densely deployed node
sensor node position estimation error and the target estimation error.An online value-iteration policy solving algorithm based on belief point is used to address the problem of high complexity of dynamic cluster-members scheduling.The algorithm realizes dynamic generation and online adjusting of sensor cluster-member nodes cooperation strategy.Simulation results show that this algorithm can improve target tracking accuracy
reduce node energy consumption and prolong network lifetime.