Fall Detection System Using Tri-Accelerometer for Wireless Body Area Network

In this paper, we observe the physical condition of elderly people or patients in personal environments such as home, office, and restroom. The elderly peoples mostly have limited physical abilities and are affected with serious physical damages even with small injuries like fall. The falls are not predictable also not avoidable. In case of a fall, its early detection with notification to emergency services is essential for quick recovery. The existing fall detection devices are more bulky and uncomfortable to wear. In this paper, a new fall detection system is used which includes tri-axial accelerometer sensors to classify the behavior and posture of the detection subject. Body motion can be classified into different patterns, i.e. vertical position, lying position, sitting position, horizontal position and fall. If a fall is suspected, an automatic message can be sent (GSM) with fall location (GPS). The major advantage of the proposed system is the use of mobile phone which is readily available to most people[1][2].