1、基于稀疏貝葉斯方法的脈間捷變頻ISAR成像技術(shù)研究來(lái)源： ： 時(shí)間：2022-01-21 【摘要】傳統(tǒng)捷變頻成像方法具有高旁瓣、低分辨率的缺點(diǎn)。鑒于捷變頻ISAR回波信號(hào)的稀疏性，該文基于原始數(shù)據(jù)的2維壓縮感知方案，在貝葉斯原理框架下，用稀疏貝葉斯算法方差成分?jǐn)U張壓縮方法(ExCoV)實(shí)現(xiàn)捷變頻ISAR像的重建。貝葉斯框架下的稀疏重構(gòu)算法考慮了稀疏信號(hào)的先驗(yàn)信息以及測(cè)量過(guò)程中的加性噪聲，因而能夠更好地重建目標(biāo)系數(shù)。作為一種新的稀疏貝葉斯算法，ExCoV不同于稀疏貝葉斯學(xué)習(xí)(SBL)算法中賦予所有的信號(hào)元素各自的方差分量參數(shù)，ExCoV方法僅僅賦予有重要意義的信號(hào)元素不同的方差分量，并擁有比SB

2、L方法更少的參數(shù)，克服了SBL算法參數(shù)多時(shí)效性差的缺點(diǎn)。仿真結(jié)果表明，該方法能克服傳統(tǒng)捷變頻成像缺點(diǎn)，并能夠?qū)崿F(xiàn)低信噪比條件下的2維高精度成像。 【關(guān)鍵詞】ISAR捷變頻壓縮感知稀疏貝葉斯學(xué)習(xí)算法方差成分?jǐn)U張壓縮方法(ExCoV) TheInterpulseFrequencyAgilityISARImagingTechnologyBasedonSparseBayesianMethod Abstract：TraditionalfrequencyagilityISARimagingmethodsuffersfromhighsidelobeandlowresolution.Toimprovether

3、esolution,byexploitingthesparsityoftargetsinthereceivedecho,thispaperusesthesparseBayesianalgorithm,namelyExpansion-CompressionVariance-componentbasedmethod(ExCoV),toreconstructtheISARimagefromtheoriginalCompressedSensing(CS)ISARdata.Bytakingintoaccountofthepriorinformationofthesparsesignalandtheadd

4、itivenoiseencounteredinthemeasurementprocess,thesparserecoveralgorithmundertheBayesianframeworkcanreconstructthescattercoefficientbetterthanthetraditionalmethods.DifferentfromtheSparseBayesianLearning(SBL)endowingvariance-componentstoallelements,theExCoVonlyendowsvariance-componentstothesignificants

5、ignalelements.ThisleadstomuchlessparametersandfasterimplementationoftheExCoVthantheSBL.ThesimulationresultsindicatethatitcanconquertheproblembroughtbytraditionalmethodsandachievehighprecisionagilityISARimagingunderthelowSNR. Keywords：ISARFrequencyagilityComressedSensing(CS)SparseBayesianLearning(SBL)algorithmExpansion-CompressionVariance(ExCoV)component 本文下載地址鏈接：/view/30ceb36052