A Solution to WSN Random Walk Model Tracking Based on Interval Clustering and Area Division

Most of traditional tracking algorithms are based on ideal models and achieve continuous position estimates and predictions by using filtering methods. However, in real-world, moving targets often have no patterns, termed as the random walk model. Hence, this paper proposes a lower-complexity solving scheme for this kind of disorder movement model. By changing positon tracking problem into a data sequence matching problem, the proposed algorithm provides a heuristic tracking scheme for random moving targets. Numerical simulation experiments show that this scheme can effectively reduce the complexity while ensuring tracking accuracy. Meanwhile, it has greater flexibility and universality.