A New Signal Denoising Method Based on Wavelet Threshold Algorithm

In order to effectively eliminate the noise in the signal, a de-noising algorithm is proposed based improved wavelet threshold algorithm. Firstly, the advantages and disadvantages of wavelet soft threshold and hard threshold method are analysis, and constructs a new threshold function with an arbitrary derivative, secondly the optimal thresholds are estimated by adjusting the parameter values, finally the simulation experiment are carried out to test the de-noising performance on the Matlab 2012 platform. The results show that, compared with other signal de-noising algorithms, the proposed algorithm can improve the signal to noise ratio, and mean square error is decreased, it can obtain a better de-noising effect and has higher practical application value.