Threshold Determination for False Track Probability in the Multipath ML-PMHT

We analyze the likelihood distribution for a multipath model in which no target is present, focusing on the peculiarities of the maximum-likelihood probabilistic multi-hypothesis tracker (ML-PMHT) when multiple peaks emerge. We study analytically the case in which peaks are separated and for the general case we propose a numerical algorithm based on sampling. The final distribution is used to compute a parameter-dependent false alarm probability threshold.

Automated summary

We analyze the likelihood distribution for a multipath model without a target and focus on the peculiarities of the ML-PMHT when multiple peaks emerge. We study the separated peaks analytically and propose a numerical algorithm based on sampling for the general case. The final distribution is used to compute a parameter-dependent false alarm probability threshold.