RBF neural network based prediction for target tracking in chain-type wireless sensor networks

A target tracking is an important embranchment of WSN, which can assure the position of a moving target real-time. This paper works on the prediction problem of target tracking of chain-type wireless sensor networks. We choose RBF neural network as the basis of the tracking prediction model. Based on analysis of chain-type tracking characters and RBF neural network based tracking prediction model, we build a target tracking prediction algorithm. The target tracking prediction problems of moving objects in coal tunnel are simulated and the simulation results show that a moving target can be traced real-time and accurately using the presented tracking prediction model and algorithm.