A Study on Wireless Sensor Node Localization and Target Tracking Based on Improved Locust Algorithm

Abstract

To improve the positioning accuracy of wireless sensor nodes and ensure the target tracking effect, a wireless sensor node positioning and target tracking method based on an improved locust algorithm is proposed. The DV Hop algorithm is used to calculate the minimum hops and average hops distance between the unknown node and each anchor node to obtain the location of the unknown node, realize the rough positioning of wireless sensor nodes, and analyze the positioning error to determine the positioning accuracy target function; The improved locust algorithm is used to solve the positioning accuracy objective function to obtain the sensor node positioning results with the minimum error; The target tracking model and the target is calculated. According to the target observation information obtained by all sensor nodes, the target state in the wireless sensor network model is tracked using the probability hypothesis density filtering algorithm. The test results show that the algorithm has better performance, the spatial evaluation index results are all lower than 0.020, and the individual distribution in the solution set is better; The location of each unknown node in different node distribution states can be obtained; The positioning error under the surface and plane is less than 0.012; The maximum error of target tracking is 0.142m; It can track single target and multiple targets.