Birth Density Modeling in the Gaussian Mixture PHD Filter in Multi-Target Tracking Problems

A recently established method for multi-target tracking is the probability hypothesis density (PHD) recursion. A closed form solution to it is provided by the Gaussian Mixture Probability Hypothesis Density filter (GM-PHD filter. Besides the GM-PHD filter algorithm implementation, choose the probability density function for representing target births in GM-PHD recursion and true target trajectory generation to get best tracking performance is a challenge and is the purpose of this paper work. One reference to judge the performance of the algorithm is the target detection time.