Human gait recognition with cane assistive device using quadratic time-frequency distributions

Abstract

In this study, the authors consider the problem of human gait recognition in the presence of a walking cane using radars. Quadratic time-frequency distributions are used to provide the local signal behaviour over frequency and to detail the changes in the Doppler and micro-Doppler signatures over time. New features that capture the intrinsic differences in the time-frequency signatures of the gait observed with and without the use of a cane are proposed. The results based on real data experiments conducted in a laboratory environment are provided that validate the effectiveness of the proposed features in discriminating gait with cane from normal human gait.