Sequential Monte Carlo and Probability Hypothesis Densities for Underwater Multitarget Tracking in Active Sonobuoy Systems

The problem of underwarter multitarget localization and tracking in active sonobuoy systems essentially belongs to the category of multi-sensor data fusion and parameter estimation. The random sets theory derived sequential monte carlo and probability hypothesis densities(SMC-PHD) filter, which can be used to obtain the approximate solution of bayesian posterior recursion for multisensor data fusion and multi-target tracking. In this paper, a clear mathematical theoretical framework and state estimation model has been constructed, and therefore the time-delay and doppler measurements obtained by each node in the active sonobuoy system are used as the input of SMC-PHD filter to output reliable estimation of unknown and time-varying number of underwater targets and their states under clutter environment. Simulation results show that the proposed method can achieve high precision tracking of time-varying number of targets in clutter environment.