On Denoising via Penalized Least-Squares Rules
2008 (English)In: 2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, 2008, 3705-3708 p.Conference paper (Refereed)
Penalized least-squares (PELS) rules for signal denoising can be obtained via the use of various information criteria (AIC, BIC, etc.) or various minmax LS approaches. Let S denote the set of "significant" parameters in the denoising problem (which is to be determined), let n(S) be the dimension of S, and let n(S)rho denote the penalty term of a PELS criterion. We show that, depending on the expression for rho, the following cases can occur: type-1) If rho does not depend on S, then denoising via the corresponding PELS rule is equivalent to simple thresholding; and type-2) If rho depends on n(S) only, then the equivalence to thresholding no longer holds but the PELS rule can still be implemented quite efficiently. We also show that the use of BIC leads to an existing PELS rule of type-1 when the noise variance in the denoising problem is known, and to a novel PELS rule of type-2 when the noise variance is unknown.
Place, publisher, year, edition, pages
2008. 3705-3708 p.
, International Conference on Acoustics Speech and Signal Processing (ICASSP), ISSN 1520-6149
signal denoising, thresholding, model order selection, information criterion
IdentifiersURN: urn:nbn:se:kth:diva-47556DOI: 10.1109/ICASSP.2008.4518457ISI: 000257456702249ISBN: 978-1-4244-1483-3OAI: oai:DiVA.org:kth-47556DiVA: diva2:455639
33rd IEEE International Conference on Acoustics, Speech and Signal Processing, Las Vegas, NV, MAR 30-APR 04, 2008
QC 201111162011-11-102011-11-102011-11-16Bibliographically approved