Research on fall detection system based on pressure sensor

This paper takes tumble as the research subject, which is an urgent accident that generally happens on elders. Considering the problems existing in current fall detection methods, we present a fall detection system based on pressure sensor, which mainly consists of pressure acquisition module, microprocessor-based data processing module, wireless communication module and mainframe. Through detecting and analyzing the plantar pressure changes under normal walking and slipping up conditions, the system realizes the real-time detection of tumble; and on the basis of this, a support vector machine based tumble identification method is put forward. Experimental results prove the reliability and veracity of the system.