Improvement of multisensor data fusion on track loss in clutter

Abstract

Improvement of multisensor data fusion on track loss in clutter is studied analytically in this paper. Calculating the transition probability density function of the fusion prediction error, the authors have analyzed the dependence of the fusion track loss statistics, such as mean time to lose track and cumulative probability of having lost track, on the clutter spatial density for nearest-neighbor association. The results show that multisensor data fusion can improve the tracking performance in clutter with low track loss probability.