自適應(yīng)多用戶檢測算法的研究.pdf

1、Y1404871分類號：UDC：TP27322密級：學(xué)校代號：10150學(xué)號：20052149峽童交彩摯碩士學(xué)位論文自適應(yīng)多用戶檢測算法的研究TheResearchofAdaptiveMultiuserDetectionAlgorithms學(xué)生姓名：導(dǎo)師及職稱：學(xué)科門類：專業(yè)名稱：研究方向：申請學(xué)位級別：論文答辯B期：學(xué)位授予單位：王培張旭秀副教授工學(xué)控制理論與控制工程智能控制碩士2007年12月20日大連交通大學(xué)大連交通大學(xué)工學(xué)碩士學(xué)

2、位論文AbstractCDMA(CodeDivisionMultipleAccess)isaprimingtechnologyforfuturemobilecommunicationsystembecauseofitsseveralfeatures、析tllwhichothermultipleaccesstechnologycallnotcompare：higherefficiencyoffrequencyallocation，robu

3、stanti—multipatheffect，macrodiversity，easyaccess，andSOonButCDMAcommunicationsystemiSainterferencelimitedsystem眥istosaythattheinterferencedirectlyinfluencesthecapacityofthissystemandtheQOSoftheuserinthesystemInordertoridt

4、hemultiaccessinterference，multiuserdetectionplaysaimportantroleItexecutesthejointlydetectionfullyusingthespreadingsequence、timedelay、phaseinformationandamplitudeItconsidersthedifferencebetweenthemultiuserinterferenceandt

5、hewhitenoise11leuplinkfrequencyresourceisusedefficiently11leperformancehasenhancedasawholeTllispaperintroducescorrespondencemultiuserexaminationprincipleandexistingmultiuserdetectionalgorithminCDMAwidefrequencyItembarksf

6、romthemultiusersignalstatisticalindependence、析mthedistributedstatisticalcharacteristicandproposestwokindofblindmultiuserexaminationsstrategies：Basedonimprovedinformationmaximizationmulti—userdetectionalgorithmandbasedonf

7、astfixed—po硫independentcomponentanalysismulti—userdetectionmethodThesetwomethodsdonotneedthetrainingsequenceandfullyusesthemultiusersignalcharacteristiccodeTheyovercometheblindseparationalgorithmontheindefinitequestionSi

8、multaneouslytheyhavealsousedthemulti—usersignalstatisticalindependenceandtheinherentcharacteristicofblindseparationalgorithmUnderunknownsituationtothesignalsource，tlleyseparateanddetecttheobjectuserUnderbigsignalandnoise

9、environment，twoalgorithmscoulddetecttargetuser謝tlllesscomputationthaninformationmaximizationalgorithmandthetraditionaldetectionalgorithmsTheyalsohavegoodperformanceinBER(biterrorrate)，convergenceandcapabilitySomesimulati

10、onsconfirmtheconclusionsTheysavethefrequencyspectrumresourcesandsimultaneouslyobtainedthebeRerseparationanddetectioneffectnlispaperalsoembarksfromthestatisticalinformationtheoryangle，inindependentparameter：KL(KullbackLei

11、bler)，twokindofblindseparationsalgorithmareunifiedunderatheoryframe，andsimultaneouslyusedintwoalgorithmsinthemultiuserexaminationtodothecomparison，nlesimulationresulthasproventhatthetheoryinferentialisreasoningandaccurac