User Posture and Movement Estimation Based on 3-Axis Acceleration with a Sensor Network

We are focusing on the development of a context-aware information platform which allows users to receive personalized information based on their context inferred by tiny sensors attached to their cellular phone. A dynamic user context inference method is one of the important technologies for realizing the context-aware services. In this poster, we show a new context inference scheme which realizes a user posture inference with only one sensor with a sensor network. To improve inference accuracy, the system automatically detects the sensor position on the user's body and selects the most relevant inference method dynamically