Characterization of Body Movement for the Real Time Fall Detection Using Wireless Sensor Module

As fall is an accident being taken place ordinarily to the elderly, it is important to prevent fractures as a result of fall by detecting fall behavior in advance. The objective of this study is to distinguish activity of daily life (ADL) from fall by using wireless sensor module. This study intends to determine fall status before the body contacts ground surface after the fall starts. In this study, natural fall and acceleration being taken place during ADL were analyzed by using tri-axial accelerometer, tilt sensor and bluetooth module so that the body will not be bound at the time of fall. Test was performed on a soft mattress so that subjects will not be injured during the test process and fall was induced through rapid movement of mattress by using a pneumatic actuator. A sensor for detecting body movement was attached to the back and waist of the subject. Fall status was determined by using acceleration value being generated from the body and direction of fall was judged by angle value. As a result of the test, in case of using this system, fall status and its direction could be correctly detected.