资源描述：
《Adaptive Radar Detection of Extended Gaussian Targets》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、AdaptiveRadarDetectionofExtendedGaussianTargetsGiuseppeRicciLouisL.ScharfUniversitàdiLecceElectricalandComputerEngineeringViaMonteroniCampusDelivery137373100Lecce,ItalyColoradoStateUniversityphone:39-0832-297205FortCollins,CO80523-1373email:giuseppe.ricci@unile.itemail:scharf@en
2、gr.colostate.eduAbstractWehaveaddressedthederivationandtheanalysisofanadaptivedecisionschemetodetectpossibleextendedtargetsmodeledasGaussianvectorsknowntobelongtoagivensubspace;noisereturnsfromthecellsundertestaremodeledasindependentandidentically-distributedGaussianvectorswitho
3、neandthesamecovariancematrix;asetofsecondarydata,freeofsignalcomponentsisalsoavailable;secondarydataareGaussian-distributedandsharethesamecovariancematrixofnoiseinthecellsundertestbutforapossibledifferentpowerlevel.Theproposeddetectorreliesonatwo-stepdesignprocedure:firstwederiv
4、etheGLRTassumingthatthenoisecovariancematrixisknownuptoascalefactor;then,wecomeupwithafullyadaptivedetectorbyreplacingthestructureofthecovariancematrixofthenoisewiththesamplecovariancematrixbaseduponthesecondarydata.Thefirststeprequiresthemaximumlikelihood(ML)estimateofthecovari
5、ancematrixoftheusefulsignal(underthesignal-plus-noisehypothesis)which,inturn,hasaknownstructure.ThatMLestimatehasbeenfirstlyproposedbyBreslerin[3];adifferentderivationisalsoproposedherein.Theperformanceassessmentisconductedresortingtothemethodproposedin[4–5]tomodelextendedtarget
6、s:thereinanexponentialmodelforfully-polarizedreturnshasbeenusedassumingthateachscatteringcentercanbecharacterizedbyits(relative)range,amplitude,andpolarizationelipse.[1]E.Conte,A.DeMaio,andG.Ricci,"GLRT-BasedAdaptiveDetectionAlgorithmsforRangeSpreadTargets,"IEEETrans.onSignalPro
7、cessing,Vol.49,No.7,pp.1336–1348,July2001.[2]T.McWhorterandM.Clark,"MatchedSubspaceDetectorsandClassifiers,"MissionResearchTechnicalReport:MRC/MRY-R-073,August2001.[3]Y.Bresler,"MaximumLikelihoodEstimationofaLinearlyStructuredCovariancewithApplicationtoAntennaArrayProcessing,"Fo
8、urthAnnualASSPWorkshoponSpectrumEstimationandMo
