IMPULSIVE NOISE REDUCTION BY USING M-TRANSFORM AND WAVELET SHRINKAGE

摘要

In this paper, the authors propose a new method for reducing both impulsive and white noise by use of M-transform and wavelet shrinkage. M-transform is a new signal transformation technique proposed by the authors, and any periodic time signal can be M-transformed into a sum of pseudorandom M-sequence. By using this property, it is shown that both impulsive noise and white noise can be M-transformed into small amplitude signals, which are removed by use of suitable filter, such as wavelet shrinkage.