Activity monitoring from real-time triaxial accelerometer data using sensor network

With the advancement in the wireless sensor network technologies, it is possible to use them for health monitoring from remote location. The monitoring of human movement can provide valuable information about individual´s daily activities. Wireless sensor node with 3-axes accelerometer were used to monitor daily activities i.e. walking, running, resting and dangerous activities such as falling. Three axes MEMS accelerometer measure body´s acceleration and transmit acceleration data to the base station attached to PC. On the PC the real time accelerometer data is processed and various inferences like walking, running and falling, are taken accordingly. This paper presents the methodologies and parameters used for chest case worn sensor as well as for wrist worn sensor and compare them. The paper also presents various design aspects of real time data acquistion from accelerometer sensor attached to sensor node. Firmware for sensor unit is developed in nesC which use tinyos as a platform. Results shows chest is better position to activity monitoring and fall detection. In total 16 experiments for chest, fall is detected in 13 with an accuracy of 81%.