Compressed Domain Contactless Fall Incident Detection using UWB Radar Signals

Falling down is one of the main reasons for hospitalization among the elderly. Constant monitoring of such vulnerable older adults and timely detection of fall incidents may significantly improve healthcare services. This paper presents a radar-based fall detection method using compressed features of the radar signals. The compressed features are obtained by using determinisitc row and column sensing. The time-frequency analysis is first performed on the radar time series and resulting spectrogram is projected onto a binary image representation. The binary images are then compressed using a 2D deterministic sensing technique by preserving the aspect ratio of the images in the compressed domain. The performance of the proposed method is evaluated using several classifiers such as support vector machine, nearest neighbors, linear discriminant analysis and decision tree. It is shown that the proposed compressive sensing based method can improve fall versus non-fall activities recognition, as evidenced by high classification metrics for low compression ratios.