Implementation of wireless sensor network based human fall detection system

Abstract

The paper presents wireless sensor architecture based human fall detection system especially for elderly people. The human fall detection system is implemented using two wireless network architectures, namely cross bow wireless sensor node system and PIC18LF4620 wireless module. The fall detection mechanism is conceived with two sensors including tri-axis accelerometer and Passive Infrared (PIR) sensor which are interfaced to monitor the activity of elderly people. The resultant voltage changes because of human activities are the targeted interest which is detected by both sensors. The real time output is compared with stored templates and the computational analysis is done within microcontroller at site itself. Any abnormality from stored templates is informed to the caretaker through wireless architecture.