An effective algorithm for dynamic pedometer calculation

The linear accelerometer is called G-Sensor. It supplies the information of accelerating change when some motion happens. The current G-Sensor products in the market are supported by 3-dimension information. The G-Sensor becomes the one of consumer components in this trend of MEMS technology development. The applications using G-Sensor components are working in the current market, for example, to be a way to control the game working, to control user interface, to drive some application programs or change slides by knocking the side of a device, to detect the falling event, and let hard disk inside the notebook have enough time to shutdown. And, G-Sensor can be implemented inside a pedometer. For wearable device, G-Sensor is used to the fitness application. The aim of this paper was to validate step counts and cadence calculations from acceleration data during dynamic activity. The proposed algorithm to calculate the accurate rate of dynamic step detection is 95%.