基于线性预测方法的水声信道最小均方误差盲均衡中英文翻译.docx

毕业设计(论文)外文资料翻译系别电子信息系专业通信工程班级B090310姓名胡月明学号B09031005外文出处J.MarineSci.Appl.附件1.原文；2.译文2013年03月BlindAdaptiveMMSEEqualizationofUnderwaterAcousticChannelsBasedontheLinearPredictionMethodAbstract:Theproblemofblindadaptiveequalizationofunderwatersingle-inputmultiple-output(SIMO)acousticchannelswasanalyzedbyusingthelinearpredictionmethod.Minimummeansquareerror(MMSE)blindequalizerswitharbitrarydelayweredescribedonabasisofchannelidentification.TwomethodsforcalculatinglinearMMSEequalizerswereproposed.OnewasbasedonfullchannelidentificationandrealizedusingRLSadaptivealgorithms,andtheotherwasbasedonthezero-delayMMSEequalizerandrealizedusingLMSandRLSadaptivealgorithms,respectively.Performanceofthethreeproposedalgorithmsandcomparisonwithtwoexistingzero-forcing(ZF)equalizationalgorithmswereinvestigatedbysimulationsutilizingtwounderwateracousticchannels.Theresultsshowthattheproposedalgorithmsarerobustenoughtochannelordermismatch.TheyhavealmostthesameperformanceasthecorrespondingZFalgorithmsunderahighsignal-to-noise(SNR)ratioandbetterperformanceunderalowSNR.1IntroductionTime-varyingcharacteristicandmulti-pathfadingofunderwateracousticchannelscaninducesevereintersymbolinterference(ISI)inhighdataratecommunicationsystems.ChannelequalizationapplyingadaptivefiltersisoneofthetechniquestomitigatetheeffectsofISI.Conventionally,theinitializationofanadaptivefilterisachievedbyaknowntrainingsequencefromatransmitterbeforedatatransmission,sothatvaluablechannelcapacityisreduced.Recently,blindequalizationtechnique(Stojanovic,1996)hasattractedmoreandmoreattention.Comparedwithadaptiveequalizationtechnique,themajoradvantageofsuchtechniqueisthatnotrainingsequenceisneededtostartuporrestartthesystemwheneverthecommunicationbreaksdownunpredictably.Traditionally,symbolratesampledchanneloutputsequenceisstationaryandhigherorderstatisticsareusedtoestimatethechannelandtocalculatetheequalizer.Morerecently,ithasbeenshownthatthechanneloutputsequenceiscyclostationaryifthesamplingrateexceedsthe