Particle filters for beam-space high resolution DOA estimation

This paper addresses the application of particle filter (PF) schemes to tracking Direction-Of-Arrival (DOA) in beam-space, using a passive array of sensors with single snapshot. This proposed technique has significant ability to estimate the DOAs of multiple sources at arbitrary times throughout the observation period, and that the requirement for quasi-stationarity over a limited interval may be relaxed. According to the State-Space Model of beam-space narrowband array signals, the importance function is established. Sampling from the importance function, the algorithm estimates the bearings of sources by sequential Monte Carlo methods. Based on DOAs estimation and the observation in each snapshot, the proposed algorithm estimates the amplitudes of the sources and the variances of noise, using a maximum posteriori estimator simultaneously. Computer simulations are presented to show the performances of the algorithms.