Human Activity Classification Based on Dynamic Time Warping of an On-Body Creeping Wave Signal

摘要

The characteristics of a nonline-of-sight on-body creeping wave are utilized to classify human motion activities. Creeping wave propagation around the torso of a nonmoving human is reviewed. Its associated complex transmission data are measured for six different motion activities on both male and female subjects. A dynamic time warping algorithm is applied on the signal magnitude and phase data for pattern recognition. The study is extended to different on-body channels and frequencies for comparisons.