Design And Research Of Multi Feature Gait Analysis System Based On Wearable Sensor

With people's attention to their own health, gait analysis is becoming a hot research area in the foot movement state. In this paper, a gait analysis system based on wearable multi feature parameters is designed, which provides a new solution for the field of gait analysis.Flexible PCB is used as electrode, carbon black and graphite dilute as conductive filler filled silicone rubber is used as sensor unit, the author designed flexible pressure sensor and tension sensor used for wearable gait analysis system, and analyzed the sensor characteristics. Eight flexible pressure sensors and a tension sensor are placed on the insole, and also a mpu-6050 inertial sensor is used. Three kinds of sensors are used to measure gait data.A data processing system for gait analysis is designed, including the microprocessor peripheral circuit module and sensor signal conditioning module and so on with CC2530 as the core. Developed the lower computer data acquisition and processing procedures, lower computer and coordinator communication program based on the ZigBee protocol, the host computer software based on the LabVIEW environment.The sensor data is sampled by the sliding time window method., and extracted the feature parameter of the data in the window. Based on the BP neural network toolbox of MATLAB software, the mapping relationship between gait characteristic parameters and gait health is established.440 training samples were selected to train the BP neural network, and the model of multi characteristic gait analysis was obtained.140 test samples were used to test the model, and the correct recognition rate was 91.4%. Test results show that the gait analysis system can realize the analysis function of human gait.