Particle filter track-before-detect algorithm based on two-hierarchy sampling

A particle filter track-before-detect (TBD) algorithm based on two-hierarchy sampling is proposed to deal with low sampling efficiency in high-dimension state space. According to each component's impacting on measurement directly or not, the higher-dimension state sampling process is converted to two successive lower-dimension state sampling processes. This algorithm improves the efficiency of state sampling for particles, and a better detection and tracking performance is obtained. Simulation results show the effectiveness of the proposed algorithm.