Bearing-only multi-target location Based on Gaussian Mixture PHD filter

Abstract

The probability hypothesis density (PHD) filter, which was derived from finite set statistics is a promising approach to multi-target tracking. An analytical closed-form solution for the PHD, named Gaussian mixture PHD Filter, is given for linear Gaussian target dynamics with Gaussian births by B. Vo and W. Ma. Based on the Gaussian mixture PHD filter, in this paper, without consideration of data association technique, a method using three passive sensors for multi-target location system is proposed, which can restrain greatly the false triangulations, calls ghosts, where the measurements of the bearing-only multi-target location system are spoiled by clutter.