A Fast Algorithm for Multiple Targets Data Association in HF Bistatic Radar System Based on MNJPDA

Abstract

Joint Probabilistic Data Association (JPDA) is one of the most effective algorithms of Multiple Targets Tracking (MTT) in dense clutter. However, with the increase of the targets tracked and the number of validation measurements, the computational cost of association probability is the choke point in engineering application. Based on the idea of Hopfield neural network applied in TSP and the characteristics of bistatic radar system, a fast algorithm for data association is developed. In this new method, the adaptive tracking gate is distributed to each target according to the location accuracy of the bistatic radar. And the posterior probability of the common measurements lying in the intersection area of target association gates, are fading down referring to a Doppler Frequency shift (DF-shift) quality factor. Simulation results demonstrate that the algorithm ensures the accuracy and real-time tracking for the data association of MTT.