Resesrch On Energy Efficient Target Tracking Algorithm For Wireless Sensor Networks

Wireless sensor networks(WSN) is an emerging network that self-organization, dynamic networking, high fault tolerance, etc., making it an extremely wide range of applications, not only can be applied to a variety of civilian areas and commercial areas, such as medical Health, environment, disaster relief, etc., can also be used in military field. In many applications of wireless sensor network, the tracking of moving targets has always been a very important application, and the research on target tracking has become one of the hot topics in wireless sensor networks.There are two core algorithms in the current research of target tracking in wireless sensor networks: localization algorithms and prediction mechanism algorithm. In the two core algorithms, the main research focus concentrated on the prediction mechanism algorithm. Base on the prediction, which nodes will be waked-up to tracking target is also a very important research point. Aiming at the problem of energy limited in Wireless Sensor Networks, a research of energy-efficient and high-accuracy object tracking algorithm is one of the most important research targets.In this paper, firstly, we discuss the value and the significance of target tracking in wireless sensor networks. And the research status and the focus of target tracking in wireless sensor networks are analyzed. Next, several key technologies are pointed out about target tracking in wireless sensor networks. What's more, it interprets and analyzes several dominant target tracking algorithms. Lastly, an energy-efficient object tracking algorithm is proposed based on the analysis and conclusions of the original researches. In the description of the algorithm, firstly, we detailedly describe the wireless sensor networks model that we proposed and the use of the two levels of perceived radius. And we design two kinds of conversion mechanism between the different states for the two types of nodes that we proposed. Secondly, according to the different tasks of the cluster headers (CH) between the phase of target detection and the phase of target tracking, we design different strategies for cluster heads election. In the phase of target tracking, judges by the induction intensity of the monitoring node, the sport trend of the target relative to the monitoring node and the remaining energy of the monitoring node itself, we set a weight value for each monitoring sensor. According to the weight value of the node, it can be determined whether the node needs to participate in tracking or not. Based on the regular hexagon clustering method in this paper, the recovery algorithm is finally proposed, which would recover the lost target with the lowest energy consumption and ultimately realize the high-energy efficiency and high-precision tracking.In order to proving the effectiveness of the energy-efficient object tracking algorithm, we simulate linear motion and curvilinear motion respectively on Matlab platform. The simulations show that selecting the greater threshold value the less energy consumption, the energy consumption of algorithm proposed by this paper gradually lower than adaptive algorithm, the network lifetime of algorithm proposed by this paper gradually longer than adaptive algorithm. Comparing to the actual target trajectory, the fitting target trajectory's error is relatively small, precision is relatively high.