File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Robust linear estimation using M-estimation and weighted L1 regularization: Model selection and recursive implementation

TitleRobust linear estimation using M-estimation and weighted L1 regularization: Model selection and recursive implementation
Authors
KeywordsImpulse Noise
Recursive Functions
Issue Date2009
Citation
Proceedings - Ieee International Symposium On Circuits And Systems, 2009, p. 1193-1196 How to Cite?
AbstractThis paper studies an M-estimation-based method for linear estimation with weighted L1 regularization and its recursive implementation. Motivated by the sensitivity of conventional least-squares-based L1-regularized linear estimation (Lasso) in impulsive noise environment, an M-estimator-based Lasso (M-Lasso) method is introduced to restrain the outliers and an iterative re-weighted least-squares (IRLS) algorithm is proposed to solve this M-estimation problem. Moreover, instead of using the matrix inversion formula, QR decomposition (QRD) is employed in the M-Lasso for recursive implementation with a lower arithmetic complexity. Simulation results show that the M-estimation-based Lasso performs considerably better than the traditional LS-based Lasso in suppressing the impulsive noise, and its recursive QRD algorithm has a good performance in online processing. ©2009 IEEE.
Persistent Identifierhttp://hdl.handle.net/10722/143326
ISSN
References

 

DC FieldValueLanguage
dc.contributor.authorZhang, ZGen_HK
dc.contributor.authorChan, SCen_HK
dc.contributor.authorZhou, Yen_HK
dc.contributor.authorHu, Yen_HK
dc.date.accessioned2011-11-22T08:30:25Z-
dc.date.available2011-11-22T08:30:25Z-
dc.date.issued2009en_HK
dc.identifier.citationProceedings - Ieee International Symposium On Circuits And Systems, 2009, p. 1193-1196en_HK
dc.identifier.issn0271-4310en_HK
dc.identifier.urihttp://hdl.handle.net/10722/143326-
dc.description.abstractThis paper studies an M-estimation-based method for linear estimation with weighted L1 regularization and its recursive implementation. Motivated by the sensitivity of conventional least-squares-based L1-regularized linear estimation (Lasso) in impulsive noise environment, an M-estimator-based Lasso (M-Lasso) method is introduced to restrain the outliers and an iterative re-weighted least-squares (IRLS) algorithm is proposed to solve this M-estimation problem. Moreover, instead of using the matrix inversion formula, QR decomposition (QRD) is employed in the M-Lasso for recursive implementation with a lower arithmetic complexity. Simulation results show that the M-estimation-based Lasso performs considerably better than the traditional LS-based Lasso in suppressing the impulsive noise, and its recursive QRD algorithm has a good performance in online processing. ©2009 IEEE.en_HK
dc.languageengen_US
dc.relation.ispartofProceedings - IEEE International Symposium on Circuits and Systemsen_HK
dc.subjectImpulse Noiseen_US
dc.subjectRecursive Functionsen_US
dc.titleRobust linear estimation using M-estimation and weighted L1 regularization: Model selection and recursive implementationen_HK
dc.typeConference_Paperen_HK
dc.identifier.emailZhang, ZG: zhangzg@hku.hken_HK
dc.identifier.emailChan, SC: ascchan@hkucc.hku.hken_HK
dc.identifier.emailZhou, Y: yizhou@eee.hku.hken_HK
dc.identifier.emailHu, Y: yhud@hku.hken_HK
dc.identifier.authorityZhang, ZG=rp01565en_HK
dc.identifier.authorityChan, SC=rp00094en_HK
dc.identifier.authorityZhou, Y=rp00213en_HK
dc.identifier.authorityHu, Y=rp00432en_HK
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.doi10.1109/ISCAS.2009.5117975en_HK
dc.identifier.scopuseid_2-s2.0-70350179784en_HK
dc.identifier.hkuros159409-
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-70350179784&selection=ref&src=s&origin=recordpageen_HK
dc.identifier.spage1193en_HK
dc.identifier.epage1196en_HK
dc.identifier.scopusauthoridZhang, ZG=8597618700en_HK
dc.identifier.scopusauthoridChan, SC=13310287100en_HK
dc.identifier.scopusauthoridZhou, Y=55209555200en_HK
dc.identifier.scopusauthoridHu, Y=7407116091en_HK

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats