A fall detector based on ultra-wideband radar sensing

Falls in the elderly have been recognized worldwide as a major public health problem. Nevertheless, falls cannot be detected efficiently yet, due to open issues on both sensing and processing sides. The most promising sensing approaches raise concerns for privacy issues (e.g., video-based approaches) or low acceptability rate (e.g., wearable approaches); whereas on the processing side, the commonly used methodologies are based on supervised techniques trained with both positive (falls) and negative (ADL-Activity of Daily Living) samples, both simulated by healthy young subjects. As a result of such a training protocol, fall detectors inevitably exhibit lower performance when used in real-world situations, in which monitored subjects are older adults. The aim of this study is to investigate a fully privacy-preserving and high-acceptance sensing technology, i.e. ultra-wideband radar sensor, together with a novelty detection methodology based exclusively on real ADL data from monitored elderly subject. The use of the UWB novelty detection methodology allowed to significantly improve detection performance in comparison to traditional supervised approaches.