The Physics of Compressive Sensing and the Gradient-Based Recovery

The Physics of Compressive Sensing and the Gradient-Based Recovery

ID:40964637

大小:1.27 MB

页数:7页

时间:2019-08-12

The Physics of Compressive Sensing and the Gradient-Based Recovery_第1页
The Physics of Compressive Sensing and the Gradient-Based Recovery_第2页
The Physics of Compressive Sensing and the Gradient-Based Recovery_第3页
The Physics of Compressive Sensing and the Gradient-Based Recovery_第4页
The Physics of Compressive Sensing and the Gradient-Based Recovery_第5页
资源描述:

《The Physics of Compressive Sensing and the Gradient-Based Recovery》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、ThePhysicsofCompressiveSensingandtheGradient-BasedRecoveryAlgorithmsQiDaiandWeiShaDepartmentofElectricalandElectronicEngineering,TheUniversityofHongKong,HongKong,China.Email:daiqi@hku.hk(QiDai);wsha@eee.hku.hk(WeiSha)ResearchReportCompiledJune8,2009Thephysicsofcompressivesensing(CS)andthegrad

2、ient-basedrecoveryalgorithmsarepresented.First,thedifferentformsforCSaresummarized.Second,thephysicalmeaningsofcoherenceandmeasurementaregiven.Third,thegradient-basedrecoveryalgorithmsandtheirgeometryexplanationsareprovided.Finally,weconcludethereportandgivesomesuggestionforfuturework.Keywords

3、:CompressiveSensing;Coherence;Measurement;Gradient-BasedRecoveryAlgorithms.c2009OpticalSocietyofAmerica1.Introduction(d)Iffissparseinthetransform-domainandthemeasurementsareacquiredinthetransform-domainThewell-knownNyquist/Shannonsamplingtheoremalso,thentheoptimizationproblemcanbegivenbythatt

4、hesamplingratemustbeatleasttwicethemax-imumfrequencyofthesignalisagoldenruleusedinminkf˜k1s.t.M0f˜=˜y.(4)visualandaudioelectronics,medicalimagingdevices,ra-dioreceiversandsoon.However,canwesimplyrecoveraFromtheaboveequations,themeaningsofthespar-signalfromasmallnumberoflinearmeasurements?Yes,

5、sitycanbegeneralized.Ifthenumberofthenon-zeroele-wecan,answeredfirmlybyEmmanuelJ.Cand`es,Justinmentsisverysmallcomparedwiththelengthofthetime-Romberg,andTerenceTao[1][2][3].Theybroughtusdomainsignal,thesignalissparseinthetime-domain.thetoolcalledCompressiveSensing(CS)[4][5][6]sev-Ifthemostimpo

6、rtantKcomponentsinthetransform-eralyearsagowhichavoidslargedigitaldatasetanddomaincanrepresentsignalaccurately,wecansaytheenablesustobuildthedatacompressiondirectlyfromsignalissparseinthetransform-domain.Becausewecantheacquisition.ThemathematicaltheoryunderlyingCSsetotherunimportantcomponents

7、tobezeroandimple-isdeepandbeautifulanddrawsfromdiversefields,butmenttheinversetransform,thetime-domainsignalcanwedon’tfocustoomuchonthemathematicalproofs.bereconstructedwithverysmallnumericalerror.TheHere,wewillgivesomephysicalexplanationsandd

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。