Gait Cycle Recognition based on Wireless Inertial Sensor Network

Abstract This paper presents a method of recognizing gait cycle based on Wireless Inertial Sensor network. The wireless inertial sensor network is embedded in shoes with accelerometers and gyroscopes to measure inertial signal during walking. In gait analysis, gait cycle is one of the most important parameters. Based on gait cycle, we can easily calculate the step continuity, step regularity, step symmetry and so on. Especially for lots of diseases estimation, such as Parkinson's disease, the accurate time of step is a very important parameter. The commercial pedometer, which can simply give the counts of gait, cannot give the accurate time of gait cycle. Based on shoes embedded with inertia sensor, we develop one method using peak detection of combined inertia sensor signal. Experimental results show the method has less error and high precision.