Research On Mobile Target Tracking In Wireless Sensor Networks

Target tracking is one of the most attractive and representative applications among the wireless sensor network. The research of how to avoid the limitation of energy resource, computing accuracy and communication distance of the sensor nodes, and take full advantage of the number of sensor nodes, topology flexibility and low communication delay is important and significant in use.The paper begin with wireless sensor network system architecture, analysis relevant methods of the physical layer to application layer and influencing factor in the target tracking applications, also research and summarize existing target tracking algorithms. What the main two innovations of this paper are:1. This paper proposes target tracking algorithm of dynamic scheduling based on the double cooperation mechanism. The algorithms systematically deal with the problem of different stages in target tracking network, and strictly distinguish the target detection node from location node, and coordinate the relationship between them, select and apply the optimum nodes. Theoretical analysis and experimental results show that this method of wireless sensor network reduces power consumption and increases the achieved positioning precision.2. This paper also proposes target tracking algorithms base on compressed sensing technology. It used the principle of node sparseness in a position of monitoring area; the complete dictionary of exploration domain is established for monitoring domain. When the mobile target achieves, the optimal matching nodes is selected by matching pursuit algorithm in dictionary. The algorithms achieve a goal of maximum and optimal nodes selection. Results of the simulation indicate that this algorithm gain optimal node location and higher position accuracy.