A wearable wireless body area network for human activity recognition

This paper presents an accelerometer-based human activity recognition system using wearable wireless body area network using ZigBee. Android mobile phone is made a part of wireless sensor network to act as a base station, which displays live information to the user. The lightweight sensor devices running Contiki and the feature of wireless design enable the users of our system to move freely. In addition, the integration with social network greatly improves the user experience. In this research, 2-axis accelerometer sensors and a threshold based algorithm are developed to recognize the human activity among standing, walking and running. The possible applications include patients motion monitoring, athletes exercise measurement and people´s daily living activity monitoring.