The research on denoising using wavelet transform

As wavelet transform is widely used in many fields, wavelet theory plays an important role in signal processing, because of it´s advantages of getting bigger output SNR and obtaining better smoothness filtering effect. First of all, the paper elaborates the advantages and fundamental concept of wavelet transform and multi-resolution. Then, it expatiates the process of signal denoising using wavelet transform. Four signals, including heavy sine, bumps, doppler and quadchirp, polluted by the noise are selected as sample data and Minimum Mean Square Error is used to go on denoising process. Then, we elaborate advantages and disadvantages of two methods, including hard threshold and soft threshold denoising, according to the comparison of denoised signal, high frequency segments and output signal-to-noise ratio. Finally, we draw the conclusion.