Injuries and complications due to falls, cause suffering both monetary and humanly. The earlier a fall is detected, the sooner actions can be taken, thus limiting the consequences. This has led to a number of products on the market today, mainly in the field of domestic monitoring.
In the project we look into the problem of solving the mobility issue, and to get basic experience of sensor networks for feature detection, (in this case gait and fall features). EISLAB is participating in two workpackages; to bridge a wireless commercial fall detection device to a cellular phone through a custom radio receiver, and to device a sensor network for data collection and processing of accelerometer data for feature detection.
The approach of reactive data processing is being investigated. The idea is to develop a distributed feature detection algorithm, that only in cases of detected feature will communicate, share features and refine calculation. This way the problem of high quality discrimination can be distributed, while the cost of communication (in terms of power and bandwidth) can be minimized.
Supervisor and contact; Per Lindgren
Research Trainee; Jimmie Wiklander
MSc. Student; David Nyström