Denosing of Rapid Variation Non-Stationary Signal Based on Wavelet Analysis

The paper aim at dealing with the problem that the telemetry rapid variation of parameter is disturbed by the noise. It propose a method based on the theory of wavelet analysis, applies the wavelet transformation of binary system to factorize the signal jammed by the noise, selects an appropriate threshold value to estimate the coefficient of the wavelet factorization, then it re-constructs the coefficient of the hige and low frequency. In this way, it realizes the separation of the signal and noise. This method is validated in the missle test data. It indicate that the wavelet transformation is an effective way to remove the noise of the non-stationary signal.