Title :
A Signal Denoising Method Based on the Improved ICI Rule
Author :
Lerga, Jonatan ; Vrankic, Miroslav ; Sucic, Victor
Author_Institution :
Fac. of Eng., Univ. of Rijeka, Rijeka
fDate :
6/30/1905 12:00:00 AM
Abstract :
In this letter, we have proposed a signal denoising method based on a modification of the intersection of confidence intervals (ICI) rule. The ICI rule is complemented by the relative intersection of confidence intervals length which is used as an additional criterion for adaptive filter support selection. It is shown that the proposed method outperforms the original ICI method equipped with the local polynomial approximation (LPA), as well as various conventional wavelet shrinkage methods.
Keywords :
adaptive filters; filtering theory; polynomial approximation; signal denoising; ICI rule; adaptive filter support selection; intersection of confidence intervals rule; local polynomial approximation; signal denoising method; wavelet shrinkage methods; Adaptive filters; Additive white noise; Bandwidth; Estimation error; Noise reduction; Polynomials; Signal denoising; Signal processing; Signal processing algorithms; Time frequency analysis; Adaptive filter support; adaptive filtering; edge preserving; intersection of confidence intervals (ICI) rule;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2008.2001817
