Multiple moving targets tracking for through-wall imaging radar

This paper mainly considers the tracking problem for multiple targets hidden behind walls using through-wall imaging (TWI) radar, and proposes an efficient multiple target tracking algorithm. Specifically, first, we present a fast wall compensation methods, based on time minimization plus look-up table technique, to obtain a high quality image. Second, the locations of the targets are extracted by two-dimensional low-pass elliptic filter and image erosion operation. Third, the measurement-to-track data association accuracy is improved through integrating nearest-neighbour (NN) data association method with amplitude information. Finally, targets' estimation state and error covariance matrix are updated by interacting multiple model (IMM) estimation algorithms. The proposed algorithm is evaluated by the real radar data.