FALL DETECTION USING WIRELESS BODY SENSOR NETWORK
MetadataShow full item record
The aim of this project is to develop a stand-alone wearable body sensor network with the primary aim of fall detection of human bodies. This network should be able to detect a fall remotely and in real time, and send an intimation along with the relevant medical details to an emergency contact and/or the nearest hospital. Several methods of fall detection have been discussed in this report and the advantages and disadvantages of them are analyzed. An optimal fall detecting mech- anism has been chosen to perform the function. Physical sensors have been used for detection of the fall, namely accelerometers and gyroscopes. Data has been collected from these sensor, and a central processing node has been utilized to process the incoming data to determine whether a fall has taken place as well as detecting the position and orientation of the body that the sensors are strapped on to. A Rasp- berry Pi processor acts as the central node. The threshold technique is employed to determine if the body has fallen over, and in what orientation it lands. The smart phone is used as a communication node and the result from the processor is obtained in a form that can be sent to the smart phone via bluetooth for further operation.