A multi-target track before detect algorithm based on CPHD for pulse Doppler radar

In strong clutter environment, the performance of traditional multi-target `track before detect' (TBD) methods on target states and target number estimation descends badly. For this reason, this paper proposed a multi-target TBD method for pulse Doppler(PD) radar based on the Cardinalized probability hypothesis density (CPHD) theory. In this method, the TBD observation model for PD radar was built firstly. Secondly, the prediction equation and update equation of an augmented dimension CPHD filter based on both Doppler and amplitude measurements was deduced by Gaussian Mixture approximating in order to estimate the targets' states and number. The target states are updated sequentially using Doppler measurements after updating them using position measurements, then more accurate state and number estimation are obtained. Simulation results show that the proposed algorithm can suppress clutter and track multi-target effectively.