Classification of Human Activities Using Variation in Impedance of Single On-Body Antenna

We propose to use the variation in input impedance of a single on-body antenna to classify human activities. The reflection coefficients of a single on-body antenna are shown to exhibit unique time-varying features for various daily activities. Employing seven human subjects, six different activities are measured by two antenna placements at three operating frequencies. A dynamic time-warping algorithm is introduced to classify the human activities. We find that the classification accuracy can be as high as 98% when the antenna is placed on the chest and operates at 433 MHz. The classification accuracy in terms of the time-window size is also presented and discussed.