Single target detection and tracking using direction-of-arrival spectrum

This paper presents a track-before-detect algorithm based on the particle filter for simultaneously detecting and tracking a single target with nearly constant velocity. The measurement data available to the algorithm is the raw direction-of-arrival spectrum of the received wideband signals at a single observation platform. The proposed method does not require prior information on the target existence probability and it is capable of detecting a target of very low signal-to-noise ratio (-30dB). Simulation results show that the proposed algorithm demonstrates good performance in detecting and tracking a target with reasonably low errors in the estimated target states over the observation period.