Application of wavelet transform in signal denoising

Abstract

The signal, which involves noise, is adverse for signal analyzing. The noise can be removed from the stationary signal by using wavelet transform or Fourier transform. But it is more advantage to remove noise from non-stationary signal with wavelet transform because of the local characteristic of the wavelet transform. Aiming at the character of the noise, this article describes, analyses, and compares several methods of removing noise from the angles of the wavelet decomposition, the wavelet reconstruction and the maximum values of the wavelet transform.