The Application of Wavelet Transformation in Signal Denoising
Sheng Guo · Tance yu kongzhi xuebao · 2002
Some property difference between signal and noise are used to restore the original signal from the polluted signal. This process depends on the aceurate description of these properties. The theory of wavelet transformation provides a novel tool for this description. In the denoising dlaorithm of Ref. , the absolute value of Corr 2 (m,n) after being normalized is compared with W(m,n) to determine whether the data point W(m,n) shoule be kept. And the standard deviation of noise at each scale is employed to stop the iterating. But this method will make coefficient produced by some signal lost in the process of correlation. So the signal can not be reconstructed perfectly. This paper tries to search the position of the signal on the small scale which corresponds to the poisition of the modulus maximum on the large scale and change the extractive manner of the signal. The simulation shows that this denoising algorithm is better than the original one