De-noising Mems Vibrating Gyro Using Wavelet Transform

In this thesis report, the theory of MEMS vibratory gyroscope is introduced and discrete wavelet decomposition and reconstruction are introduced. In the first part of this work, I review the Coriolis effect and angular rate sensors, and fundamental operational principles of micro-machined vibratory gyroscopes. The MEMS gyroscopes are expected to lead to reliable, robust and high performance angular-rate sensors with low production costs and high yields, fitting into or enabling many applications in the aerospace/defense, automotive and consumer electronics markets. Then the method of de-noising of the noised signal using wavelet decomposition and thresholding method are described. Noise estimation, threshold selection methods are also given. The specifications of the Rate Sensor Gyro – RRS01-03-0100 are given in the third chapter. At last, the experimental results are discussed. The experimental results show that the method presented can greatly reduce the noise in the gyro‟s output signal and improve the performance of the system. In this work , the data analysis approach to be described can be understood as a transform which maps a hierarchical clustering into a transformed set of data; and this transform is invertible, meaning that the original data can be exactly reconstructed. Such transforms are very often used in data analysis and signal processing because processing of the data may be facilitated by carrying out such processing in transform space, followed by reconstruction of the data in some “good approximation” sense.