Performance of ZigBee Based Fall Detection Alarm System

Mohamed Hadi HABAEBI  Mabruka Mohamed AGEL  Alhareth ZYOUD  

IEICE TRANSACTIONS on Communications   Vol.E99-B   No.2   pp.385-391
Publication Date: 2016/02/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.2015EBP3157
Type of Manuscript: PAPER
Category: Network
wireless sensor networks,  accelerometers,  fall detection,  ZigBee standard,  

Full Text: PDF(2.8MB)
>>Buy this Article

Accidental falling among elderly people has become a public health concern. Thus, there is a need for systems that detect a fall when it happens. This paper presents a portable real-time remote health monitoring system that can remotely monitor patients' movements. The system is designed and implemented using ZigBee wireless technologies, and the data is analysed using Matlab. The purpose of this research is to determine the acceleration thresholds for fall detection, using tri-axial accelerometer readings at the head, waist, and knee. Seven voluntary subjects performed purposeful falls and Activities of Daily Living (ADL). The results indicated that measurements from the waist and head can accurately detect falls; the sensitivity and reliability measurements of fall detection ranged between 80% and 90%. In contrast, the measurements showed that the knee is not a useful position for the fall detection.