Human activity classification and localization algorithm based on temporal-spatial virtual array

摘要

We propose a human activity classification and localization algorithm based on temporal-spatial virtual array. By analyzing the synthetic wave of the echo reflected from each human body part, we can classify various human activities. In addition by using the estimation result of target direction with temporal-spatial phase shift that depends on relative target velocity and position, the proposed algorithm can estimate the position of a human being simultaneously with a smaller amount of computational complexity than the conventional least-square approach. We demonstrate the effectiveness of the proposed algorithm via simulation results.