On the Detection of Indoor Falls Using an Ultra-Wideband Radar

The present study aims to assess the feasibility of an ultra-wideband radar to distinguish between typical bathroom activities and simulated falling incidents. A threshold-based algorithm using the relative change of power received by the radar was used to identify the presence of an individual in the laboratory setting and whether a fall had occurred. Results showed that the proposed fall detection algorithm exhibits robust performance in terms of true and false positive rates, yielding 96.7% and 5%, respectively.