Motion Mode Recognition and Step Detection Algorithms for Mobile Phone Users

摘要: Microelectromechanical Systems (MEMS) technology is playing a key role in the design of new generation smartphones. Thanks to their reduced size, power consumption, MEMS sensors can be embedded above mobile devices for increasing functionalities. However, cannot allow accurate autonomous location without external updates, e.g., from GPS signals, since signals are degraded by various errors. When these fixed on user's foot, stance phases foot easily determined and periodic Zero velocity UPdaTes (ZUPTs) performed bound position error. sensor hand, situation becomes much more complex. First all, hand motion decoupled general user. Second, characteristics inertial differ depending carrying modes. Therefore, algorithms characterizing gait cycle pedestrian using handheld device have been developed. A classifier able detect modes typical phone users has designed implemented. According detected mode, adaptive step detection applied. Success process found higher than 97% all