Using Translation Invariant Discrete Wavelet Transform to Improve Bias Stability of Fiber Gyroscope

摘要

The method using translation invariant discrete wavelet transform arithmetic to improve gyroscope bias stability is described. On the basis analyzing the noise characteristic of fiber gyroscope signal, the suitable wavelet vector was selected to extract the noise from signal. The TIDWT, which can avoid the Pseudo-Gibbs phenomena, had be adopted to construct the denosing arithmetic, then the arithmetic was used to filter the real gyroscope signal, the result indicated that the bias stability was improved about 0.6% effectively.