MEMS gyro signal de-noising method based on extended recursive least square

This paper discusses the design of extended recursive least square method based on time series analysis in order to overcome large noise and low precision of MEMS gyro. The method adopts the forgetting factor-based recursive least square which can work well even with uncertain noises. Firstly, ARMA models are used to model gyro random drifts. Secondly, variable forgetting factors enhance the robustness of extended recursive least square approach. Some experiments are carried out and the results show that the proposed method advances the performance of MEMS gyro signal de-noising. It gains better accuracy and better robustness than traditional Kalman filter.