A Smartphone Location Independent Activity Recognition Method Based on the Angle Feature

摘要: The smartphone-based human activity recognition method is helpful in the context awareness, health monitoring and inertial positioning. Comparing with traditional wearable computing which fixes accelerometers on specific positions of a user body, based smartphone faces problem varying sensor locations. In this paper, we lay emphasis study feature extraction algorithm independent phone First, angle motion model presented to illustrate activities. describes difference among walking, going upstairs downstairs. Then, an proposed according model. Our analysis shows that different activities have significantly features. Finally, our experiments are introduced. include data collecting, results. results show accuracy improved by 2% through adding original