Distributed intelligent architecture for falling detection and physical activity analysis in the elderly

This is the authors' abstract. We don't add key points for this paper.

A novel approach for the detection of falls, the analysis of body postures, mobility and metabolic energy expenditure of elderly people has been developed. It is based on a distributed intelligence architecture, supported by it wireless personal area network (WPAN) which allows a full 24-hour supervision of the user, both indoor and outdoor home. An open design methodology lets the addition of new sensors for the on-line monitorization or other biosignals. In this paper general guidelines and design issues are reported, with special emphasis on the Intelligent Accelerometer Unit (IAU), based on a four-axis accelerometer, the signals of which are transmitted to the WPAN server (PSE) for on-line processing. The availability of three axis in the median plane provides an inclination measurement with high sensibility. The IAU can be worn like a patch, fixed to the back, at the height of the sacrum. A prototype of the IAU is currently under validation phase, in order to optimize signal transmission between IAU and PSE.