A Linear Complexity Approximate Method for Multi-Target Particle Filter Track before Detect

Abstract

The particle filter offers the optimal Bayesian filter for track before detect with a single target. However, direct application to the case of multiple targets can be infeasible because the number of particles required grows exponentially. This paper presents a new method for efficiently implementing track before detect for multiple targets using particles. This method is compared with alternative options on a challenging scenario with up to 20 targets.