Rao and Wald Tests for Nonhomogeneous Scenarios 英文参考文献.docVIP

Sensors2012,12,4730-4736;doi:10.3390/s120404730 OPENACCESS sensors ISSN1424-8220 /journal/sensors Article RaoandWaldTestsforNonhomogeneousScenarios ChengpengHao *,DaniloOrlando andChaohuanHou 1, 2 1 1 TheStateKeyLaboratoryofAcoustics,InstituteofAcoustics,ChineseAcademyofSciences, Beijing100190,China ElettronicaS.P.A.,ViaTiburtinaValeria,Rome00131,Italy 2 *Authortowhomcorrespondenceshouldbeaddressed;E-Mail:haochengp@; Tel.:+86-10-8254-7719;Fax:+86-10-8254-7705. Received:12March2012;inrevisedform:28March2012/Accepted:29March2012/ Published:12April2012 Abstract: Inthispaper,wefocusonthedesignofadaptivereceiversfornonhomogeneous scenarios.Moreprecisely,atthedesignstageweassumeamismatchbetweenthecovariance matrix of the noise in the cell under test and that of secondary data. Under the above assumption, weshowthattheWaldtestistheadaptivematched?lter, whiletheRaotest coincideswiththereceiverobtainedbyusingtheRaotestdesigncriterioninhomogeneous environment, hence providing a theoretical explanation of the enhanced selectivity of thisreceiver. Keywords:nonhomogeneousscenarios;Raotest;Waldtest;mismatch 1.Introduction InrecentyearsadaptivedetectionoftargetsembeddedinGaussiandisturbancewithunknownspectral properties has received an increased attention in the signal processing community. In the seminal paperbyKelly[1],thegeneralizedlikelihoodratiotest(GLRT)isusedtodesignanadaptivedecision schemecapableofdetectingasignalknownuptoascalingfactorinpresenceofGaussiandisturbance withunknowncovariancematrix. Moreover,itisassumedthatasetofsecondarydatafreeofsignal components,butsharingthesamespectralpropertiesofthenoiseinthecellundertest(CUT),isavailable (homogeneous environment). In [2] the authors derive the so-called adaptive matched ?lter (AMF), whosedesignreliesontheso-calledtwo-stepGLRT-baseddesignprocedure. Infact,itsdesignissplit  Sensors2012,12 4731 intotwosteps:?rstanon-adaptiveGLRTforknowncovariancematrixisderived;then