稀疏傅里葉變換,Sparse Fourier Transform,FFT的革命性演進(jìn)_第1頁
稀疏傅里葉變換,Sparse Fourier Transform,FFT的革命性演進(jìn)_第2頁
稀疏傅里葉變換,Sparse Fourier Transform,FFT的革命性演進(jìn)_第3頁
稀疏傅里葉變換,Sparse Fourier Transform,FFT的革命性演進(jìn)_第4頁
稀疏傅里葉變換,Sparse Fourier Transform,FFT的革命性演進(jìn)_第5頁
已閱讀5頁，還剩25頁未讀， 繼續(xù)免費閱讀

下載本文檔

版權(quán)說明：本文檔由用戶提供并上傳，收益歸屬內(nèi)容提供方，若內(nèi)容存在侵權(quán)，請進(jìn)行舉報或認(rèn)領(lǐng)

文檔簡介

TheSFT:

FromTheorytoPracticeHistoryWhatishappening……WorkshoponSparseFourierTransformDate:

17-18February,2013

Location:

MIT

15talkscoveringtopicsfromtheorytoapplication(includingphotography,arrayprocesing,sepctralcompressivesensing,FPGA-baseddesign....)DFT&FFTTimecomplexity:O(n2)FFTisan

algorithm

tocomputetheDFTTimecomplexity:O(n*logn)SparseFourierTransformGivenacomplexvectorxoflengthn,andaparameterk,estimatetheklargestcoefficientsoftheFouriertransformofx.Inmanyapplications,mostoftheFouriercoefficientsofasignalaresmallorequaltozero.SparseFourierTransformSFTworksbybinningtheFouriercoefficientsintoasmallnumberofbuckets.Duetothesparsity,eachbucketislikelytohaveonlyonelargecoefficient,whichcanbelocated(tofinditsposition)andestimated(tofinditsvalue)Usingann-dimensionalfiltervectorGthatisconcentratedbothintimeandfrequencypassregionOnecanrandomizethepositionsofthefrequenciesbysamplingthesignalintimedomainappropriatelySparseFourierTransformEstimation：thephasedifferencebetweentwosamplesofthefilteredsignalislinearintheindexofthecoefficient,andhencewecanrecovertheindexbyestimatingthephasesUpdatingthesignal:thefilteringprocessneedstoberepeatedtoensurethateachcoefficientiscorrectlyidentifiedOngoingsFFTProjects(BeyondTheory)LightFieldPhotographySpectrumSharingMedicalImagingGPSsFFTChipSpectrumCrisisTheFCCpredictsaspectrumcrunchstarting2013Butatanytime,mostofthespectrumisunusedSpectrumSharingSensetofindunusedbands;Usethem!HowdoyoucaptureGHzofspectrum?SeattleJanuary7,2013ChallengesinSparseGHzAcquisitionGHzsamplingisexpensiveandhigh-powerTensofMHzADC<adollarLow-powerAFewGHzADCHundredsofdollars10xmorepowerCompressivesensingusingGHzanalogmixingisexpensive,andrequiresheavycomputationHashthespectrumintoafewbuckets

Estimatethelargecoefficientineachnon-emptybucket

RecapofsFFT1-Bucketize

2-EstimateCanignoreemptybucketSpectrumSensing&DecodingwithsFFTBucketizeEstimateSpectrumSensing&DecodingwithsFFTBucketizeEstimateSub-samplingtimeAliasingthefrequenciesSpectrumSensing&DecodingwithsFFTHashfreqs.usingmultipleco-primealiasingfiltersSamefrequenciesdon’tcollideintwofiltersIdentifyisolatedfreq.inonefilterandsubtractthemfromtheother;anditerate…BucketizeEstimateLow-speedADCs,whicharecheapandlow-powerSpectrumSensing&DecodingwithsFFTEstimatefrequencybyrepeatingthebucketizationwithatimeshift?TBucketizeEstimate

BigBand:Low-PowerGHzReceiverBuilta0.9GHzreceiverusingthree50MHzsoftwareradiosFirstoff-the-shelfreceiverthatcapturesasparsesignallargerthanitsowndigitalbandwidthConcurrentSendersHoppingin0.9GHzNumberofMHzSendersRandomlyHoppingginin0.9GHzRealtimeGHzSpectrumSensingCambridge,MAJanuary2013sFFTenablesaGHzlow-powerreceiverusingonlyafewMHzADCsProbabilityofDeclaringaUsedFrequencyasUnusedOngoingsFFTProjects(BeyondTheory)LightFieldPhotographySpectrumSharingMedicalImagingGPSsFFTChipMagneticResonanceSpectroscopyAnalysesthechemicalmakingofabrainvoxelDiseaseBio-markersChallengesLongacquisitiontimepatientisinthemachinefor40mintohoursArtifactsduetoacquisitionwindowWindowingArtifactsFouriertransformofawindowisasinc(Inverse)FourierTransformAcquisitionWindowConvolutionwithasincWindowingArtifacts

ConvolveConvolveDiscretizationDiscretizationTailChallengeswithIn-VivoBrainMRSclutterduetosinctailhoursinmachineCansparserecoveryhelp?CompressiveSensing+30%dataLostsomeBiomarkersNon-IntegerSparseFFTProblemandModelSparseinthecontinuouscaseTherailingsarebecauseofnon-integerfrequenciesAlgorithmUseoriginalsparseFFTtoestimateintegerfrequenciesUsegradientdescentalgorithmtofindthenon-integerfrequenciestominimizetheresidueofourestimationoverthesamplesChallengeswithIn-VivoBrainMRSclutterduetosinctailhoursinmachineCansparserecoveryhelp?SparseFFT+30%ofdataRemovedClutterwithoutlosingBiomarkerssFFTprovidesclearerimageswhilereducingtheacquisitiontimeby3xLight-FieldPhotographyGeneratedepthandperspectiveusingimagesfroma2DcameraarrayImagesarecorrelated4DfrequenciesaresparseGoal:

SameperformancebutwithfewerimagesOriginalReconstructedwith11%ofdataConclusionManyapplicationsaresparsein

溫馨提示

  • 1. 本站所有資源如無特殊說明，都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請下載最新的WinRAR軟件解壓。
  • 2. 本站的文檔不包含任何第三方提供的附件圖紙等，如果需要附件，請聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
  • 3. 本站RAR壓縮包中若帶圖紙，網(wǎng)頁內(nèi)容里面會有圖紙預(yù)覽，若沒有圖紙預(yù)覽就沒有圖紙。
  • 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
  • 5. 人人文庫網(wǎng)僅提供信息存儲空間，僅對用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理，對用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯，并不能對任何下載內(nèi)容負(fù)責(zé)。
  • 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容，請與我們聯(lián)系，我們立即糾正。
  • 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時也不承擔(dān)用戶因使用這些下載資源對自己和他人造成任何形式的傷害或損失。

評論

0/150

提交評論