An improved dynamic programming algorithm for target detection and tracking in non-homogeneous clutter

The performance of dynamic programming (DP) based track-before-detect (TBD) algorithm is seriously affected by non-homogeneous clutter. A novel approach which combines DP algorithm with priori information is proposed in this paper. The method of choosing the valid state transitions reasonably according to the priori information of clutter intensity is analysed. And the compensation strategy to merit function is developed. Simulations demonstrate the meaningful advantage that this improved DP algorithm ensures better target tracking performance than traditional DP process in non-homogeneous clutter.