Unsupervised Learning-Based Unobtrusive Fall Detection Using FMCW Radar

It is necessary to detect the fall of the elderly in time. As a non-contact monitoring device, radar can monitor users without their knowledge and protect their privacy. The unsupervised fall detection method does not need to collect and label fall samples, which avoids the difficulty of collecting fall data and saves researchers time and cost. The current unsupervised fall detection studies consider fewer types of actions and do not test the generalization of their models in new environments and subjects. This paper proposes a new unsupervised fall detection system, including a feature extractor and predictor. We first use 3D convolution and 3D transposed convolution to construct a feature extractor to extract the range-velocity-time features of radar signals. Then, we construct a predictor to learn the pattern of non-fall action. Finally, we design an unsupervised training method based on hard sample mining to improve the ability of the model to identify hard negative samples. We train the model using only unlabeled non-fall samples and test it in new scenarios. The system's accuracy in the data set containing 52 kinds of non-fall actions and 12 kinds of fall actions is 95.54%, the false alarm rate is 1.07%, and the AUROC is 0.9974.