The Enhanced Probability Hypothesis Density-based Filter for Multitarget Tracking and Counting

Abstract

Efficient multiple target tracking and counting have become an essential requirement for many wireless binary sensor networks applications. This paper investigates the problem of tracking and counting multiple individual targets that are present in a binary sensor network. An enhanced probability hypothesis density-based filter is proposed by introducing the spatial and temporal dependencies in order to improve the targets localization accuracy. This paper investigates four of the existing target tracking and counting algorithms: 1) ClusterTrack filter, 2) A Distributed energy efficient algorithm (DEE), 3) Multicolor particle filter technique (MCPF) and 4) Probability hypothesis density filter. The implementation of dynamic counting techniques is considered to improve the efficiency of the estimations of targets trajectories. Simulations compare the performance of the proposed algorithm with the previously mentioned target tracking approaches, to verify the efficiency and accuracy of the proposed target counting and tracking technique in binary sensor networks.