《Random Matrix Theory Based Cognitive Radio Spectrum Sensing》.pdf

JournalofCommunicationandComputer11(2014)312—322 Random MatrixTheoryBasedCognitiveRadioSpectrum Sensing usamaY．MohamadandDirkDahlhaus CommunicationsLaboratory，UniversityofKassel,WilhelmshoherAllee73，KasselD-3500,Germany Received：January19，2014 ／Accepted：February 12，2014 ／Published：M arch31，2014 Abstract：A cognitiveradios3stem allowshigherdatatransmissionratesduetotheeffi cientspectrum utilization．Specturm sensing playsasubstna tialroleinsuch acognitionscenario．Inhtispaper，anovelmultiplenatennasensingalgorithm isproposed for detectinghtepresenceorhteabsenceofhteprimaryusersing a1．TheschemeiscalledCRABWISE fCognitiveRAdiosensingBased onmeiointdistributionofpseudoWlSha~matrixEigenvalues)．ItturnsouthtatwithoutpriorinofrmationabouthtePU (primary user)singal，CRABWISEperformsneartotheoptimalsensingperformance，whichisobservedforanenergydetectionsensingbenig equippedwiht perfectpriorinfomr ation ofthePU sing a1．TheperformanceOfCRABW ISE isinvestigatedusing thereceiver operatingcharacteristicofrsingalstransmittedoveradelay．dispersivechann e1．Moreover，wesutdyhow to findhteoptimum thresholdofrmeproposedtestnmu erically．Theachievableperfomr anceisconsideredofrnicreasinglengthofhtereceivedsing al flameintemr sofbOthprobabilityofdetectionandprobabilityofawrongdecision． Keywords：PseudoW ishartmartix，Bayesiantest，likelihoodratio，receiveroperatingcharacteristic 1．Introduction 儿l_dI儿l I． Although there isno clear-cutdefinition ofrCR TomeettheeVe卜increasing demandon high data system，therearesomekeycomponentsinthissystem． rates expected by the International Thelattercomprise，ofrinstance，spectrum sensingor TelecommunicationUnion[1]especiallyformobile e