LAUSR

Most of the traditional tracking algorithms use the Kalman filter to predict the tracking process. Although the tracking accuracy is relatively high, the calculation is large and the time complexity is high. Based on this research, this paper proposes a dynamic clustering target tracking algorithm for motion trends. The algorithm forms a dynamic cluster in the network, and the cluster head dynamically schedules the nodes to achieve collaborative tracking of the targets. The tracking strategy is mainly divided into two stages: First, the cluster head establishes a “neighbor node set” within its communication range, and selects the neighbor node in the “neighbor node set” according to the distance between the node and the target to construct the “intra-cluster member set” to perform the target on the target. Tracking; as the target moves continuously, the cluster head updates the members in the cluster at regular intervals, removes the nodes that have lost the target monitoring from the cluster, and adds the new nodes to the cluster; secondly, elects a new cluster head; if current When the cluster head is no longer suitable to continue to serve as the cluster head, the current cluster head selects the node in the “intra-cluster member set” as the new cluster head of the next work cycle; according to the moving direction of the target, selects the node with the best moving tendency of the target For the new cluster head, this allows the new cluster head to have a longer duty cycle and avoid frequent replacement of the cluster head; the new cluster head continues to set up the dynamic cluster to track the target until the target moves out of the monitoring area. The simulation results show that the proposed algorithm is more efficient than the traditional target tracking method.