Title :
Impulsive noise elimination using polynomial iteratively reweighted least squares
Author :
Kuruoglu, Ercan E. ; Rayner, Peter J W ; Fitzgerald, William J.
Author_Institution :
Dept. of Eng., Cambridge Univ., UK
Abstract :
A new nonlinear filtering technique is introduced for the elimination of impulsive noise modelled with a symmetric α-stable (SαS) distribution. The new algorithm, called polynomial iteratively reweighted least squares (PIRLS), employs a Volterra filter the coefficients of which are estimated by minimizing the lp-norm of the estimation error. The filter, hence constructed, is used to estimate the clean data from the corrupted data. Simulation results obtained for audio data corrupted by synthetic SαS noise indicate that PIRLS is very successful in removing impulsive noise
Keywords :
Volterra equations; acoustic signal processing; approximation theory; audio signals; error analysis; filtering theory; interference suppression; iterative methods; least mean squares methods; nonlinear filters; polynomials; statistical analysis; Volterra filter; audio data; clean data estimation; corrupted data; estimation error; filter coefficients; impulsive noise elimination; nonlinear filtering; polynomial iteratively reweighted least squares; simulation results; symmetric α-stable distribution; synthetic SαS noise; Acoustic noise; Electromagnetic interference; Filtering; Gaussian noise; Least squares approximation; Least squares methods; Polynomials; Signal processing; Signal processing algorithms; Wiener filter;
Conference_Titel :
Digital Signal Processing Workshop Proceedings, 1996., IEEE
Conference_Location :
Loen
Print_ISBN :
0-7803-3629-1
DOI :
10.1109/DSPWS.1996.555532
