Wavelet thresholding using higher-order statistics for signal denoising

The paper demonstrates a higher-order statistics (HOS) based method of wavelet thresholding for signal denoising. We calculate the triple correlation coefficients of wavelet-signal correlation for identification of wavelet coefficients uncorrupted by noise. Since the higher than second-order moments of the Gaussian probability function are zero, the Gaussian noise can be eliminated completely. The method is also valid for unknown spectral density noise. The results of computer simulation show the availability and the effectiveness of the proposed wavelet thresholding method.

[1]  Georgios B. Giannakis,et al.  Signal detection and classification using matched filtering and higher order statistics , 1989, IEEE Trans. Acoust. Speech Signal Process..

[2]  David L. Donoho,et al.  De-noising by soft-thresholding , 1995, IEEE Trans. Inf. Theory.

[3]  I. Johnstone,et al.  Wavelet Shrinkage: Asymptopia? , 1995 .

[4]  I. Johnstone,et al.  Ideal spatial adaptation by wavelet shrinkage , 1994 .

[5]  A. Lohmann,et al.  Triple correlations , 1984, Proceedings of the IEEE.

[6]  Stéphane Mallat,et al.  Singularity detection and processing with wavelets , 1992, IEEE Trans. Inf. Theory.

[7]  Dennis M. Healy,et al.  Wavelet transform domain filters: a spatially selective noise filtration technique , 1994, IEEE Trans. Image Process..

[8]  Stéphane Mallat,et al.  Characterization of Signals from Multiscale Edges , 2011, IEEE Trans. Pattern Anal. Mach. Intell..