A Data Processing Method of Silicon Micro Structural Sensors Based on H-infinity Filtering Technology

Aiming at the problem that the data of the low cost silicon micro structural sensors is difficult to be adopted immediately because of the existing disturbance, a robust filtering method is proposed in this paper. Because the disturbance signal is not the white Gaussian noise and the statistical properties of the noise and system is not well known, it is failure to the traditional Kalman filtering. Therefore, we use the H∞ filtering technology to solve the problem by constructing a twice-integral model of the sensors. The results of experiments indicate that this method is simple and effective and can remove the noise from the real-time signals effectively.