A Method with Triaxial Acceleration Sensor for Fall Detection of the Elderly in Daily Activities

Falls are one of the major risks which the elderly people face. Recently, due to the demands for guardianship of physical functions, the device to detect falls automatically is urgently needed. This study was focused on the method of fall detection and the wireless device based on a triaxial acceleration sensor. To evaluate the performance, experiments were conducted on fall, squat, stand up and walk. The device was also set in three body positions (head, shoulder and belly) to get fall signals compared. It is considered that the difference between valley and peak of the acceleration on z axis can be used to detect falls as an obvious feature. If it's higher than 0.5V, it can be concluded that the person has a fall occurrence. The device is expected to be useful to detect falls of the elderly as healthcare equipment.