Dynamic Programming Implementation of Track-Before-Detect Algorithm

This article presents the implementation of track-before-detect using dynamic programming algorithms. For this implementation, we configured the simulation environment mainly referencing the target measurement model and scenario setup from D.J. Salmond's "A Particle Filter for Track-Before-Detect" paper. The dynamic programming algorithm implementation follows the approach described in Dr. Yi Wei's doctoral dissertation from University of Electronic Science and Technology of China. The algorithm employs state transition matrices for target motion prediction and calculates cumulative merit functions recursively across time steps. Through these references, we successfully implemented the track-before-detect algorithm with reliable results. The implementation features recursive cost calculation and optimal path backtracking mechanisms that efficiently handle measurement uncertainty. We believe this algorithm shows strong potential for practical applications and can provide substantial support for research and applications in related fields.