Association of range-doppler video sequences in multistatic UWB radar for human tracking

Novel target association approach for indoor human tracking is designed for the multistatic ultra-wideband (UWB) radar. A new concept, namely range-Doppler video sequence, is proposed to describe the slow-time evolution of human movement in range-Doppler domain. We apply the video time density function to model human movement dynamics of range-Doppler video sequences in slow time. Sliding window correlation algorithm is developed to extract the movement rhythm of personnel in different receiving channels. Experimental results demonstrate that the same walking rhythms were obtained in different channels, which proves that our approach is capable of solving the multiple human targets association problem for multistatic UWB radar.