An MCMC-BASED PARTICLE FILTER TBD ALGORITHM FOR MULTI-TARGET DETECTION AND TRACKING

Abstract

This paper presents an MCMC-based particle filter TBD algorithm for multi-target detection and tracking in low SNR environment.In this algorithm, a set of unweighted samples is used to represent the posterior distribution of the joint multi-target state.Then, a MH algorithm is adopted to recursively obtain the unweighted samples approximation for the posterior.Simulation results show that the performance of the proposed algorithm is very well for a fix known number of targets, even targets' paths are close proximity.