Gaussian mixture implementation of PHD filter based on entropy distribution

As far as the probability hypothesis density filter is concerned, a Gaussian mixture implementation based on entropy distribution is proposed. Entropy distribution is adopted as the prior distribution of mixture parameters in the algorithm. Entropy distribution is applied to pruning the irrelevant mixture components during the iteration of maximum a posterior. The pruning operation is done by adjusting the mixing weights. Besides, the problem that one intensity peak is described by several mixture components with similar parameters, can be solved by using the proposed algorithm. Simulation results show that the proposed algorithm is superior to the typical threshold algorithm.