Research On Target-tracking Algorithms In Wireless Sensor Network

Wireless sensor network, which is made by the convergence of sensor, nested computation, networks and wireless communication technologies, is a novel technology about acquiring and processing information. It can be used for sensing, collecting and processing information of monitored objects. Target tracking is one of the focus issues in WSNs.In this paper, firstly, we discuss the value of target tracking in wireless sensor network. And the research status and the focus of wireless sensor network are analyzed. Next, several key technologies and guide line are pointed out about target tracking in wireless sensor network. What's more, it analyzes and compares several dominant target-tracking algorithms. Lastly, a new target-tracking algorithm based on dynamic cluster (DCTA) is proposed. It improves the disadvantages of localization and tracking in DPT.DCTA algorithm uses a relatively energy-efficient dynamic cluster network model. Dynamic cluster with the target trajectory wakes up one-hop nodes to track the target, which cannot blindly wake up nodes to waste unnecessary energy. According to the signal strength of each node accepting from the target, cluster headers (CH) weigh to the information to more accurately localize and predict the target. Moreover, the paper detailedly analyses the reasons of the target losing. Though the geometry, it is proved that the target must be sensed when there are six sensors in the circular monitoring area. According to the conclusion, the paper works out a math interval that can recover the missing target if the current CH wakes up the next CH during the interval. As the self-organization of wireless sensor networks and the specialization of the application environment, network is not relatively steady like ordinary networks for nodes death. In these instances, the algorithm also detailedly considers the exception in the process of tracking the target and solutions.Finally, the DCTA algorithm is simulated on Matlab platform. The simulations show that the half-time of the DCTA network model is nearly 27.3% higher than grid, the localization algorithm is clearly improved, the precise rate of tracking algorithm achieves 0.942, and in the same speed tracking the missing rate of DCTA algorithm is 26.8% lower than DPT algorithm. With the speed increasing, the energy consumption of DCTA algorithm is gradually lower than DPT algorithm.