A New Threshold Method Exploration about Wavelet De-noising

In order to improve the de-noising effect of wavelet threshold further, after analyzing the shortcomings of the traditional soft and hard threshold de-noising methods, this paper proposes a trade-off of soft and hard threshold method. The typical blocks signal superposed standard Gaussian white noise in the Matlab wavelet toolbox is taken as an example for simulating and testing, to verify the feasibility and availability of this new threshold method. Signal-to-noise ratio (SNR) and root-mean-square error (RMSE) are taken as the appraisal parameters of de-noising performance. The experiment results indicate the superiority of the new threshold method, and it improves the de-noising effect to some extent.