A study of detection of trip and fall using Doppler sensor on embedded computer

A recent social problem in Japan is domestic accidents involving elderly persons, who are increasingly living alone. Domestic accidents are unpredictable emergent situations, yet the constant care of elderly person by non-relatives or caregivers is difficult due to associated labor costs, privacy problems, and burdens placed on elderly persons themselves. This paper targets trip and fall, the leading cause of accidental death in elderly persons, and examines a detection method for such accidents. To address privacy issues, we use a Doppler sensor that can detect movement without the use of personally identifying data. We also investigate the movement detection and identification by frequency analysis of sensor output using fast Fourier transform. In addition, we design and implement a prototype terminal to detect trip and fall by using a Doppler sensor and an embedded computer.