Fall detection is an area of increasing interest in independence-assisting remote monitoring technologies for the elderly population. Immediate assistance following a fall can lower the risk of medical complications, thus saving lives and reducing the associated health care costs. Therefore it is important to detect a fall as it happens and promptly mobilise first responders for proper care and attendance to possible injury. Radar offers privacy and non-intrusive monitoring capabilities. Micro-Doppler signatures are typically employed for radar-based human motion detections and classifications. Proper time–frequency signal representation is, therefore, required from which important features can be extracted. Missing or noise/interference corrupted data can compromise the integrity of micro-Doppler signatures and subsequently confuse the classifier. In this study, the authors restore the time–frequency signatures associated with human motor activities, such as falling, bending over, sitting and standing, by using a hybrid approach of compressive sensing and multi-window analysis based on Slepian or Hermite functions. Because time–frequency representations of many human gross-motor activities are sparse and share common support in joint-variable domains, the multiple measurement vector approach can be effectively applied for fall classification in both cases of full data or compressed observations.