预览加载中，请您耐心等待几秒...

基于实虚型连续多值复数Hopfield神经网络的QAM盲检测.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于实虚型连续多值复数Hopfield神经网络的QAM盲检测 1.Introduction QAM(QuadratureAmplitudeModulation)isawidelyusedmodulationschemeinmoderncommunicationsystems.QAMmodulationallowsforthetransmissionofmultiplebitspersymbolandthusitisanefficientuseofbandwidth.However,QAMsignalsarevulnerabletonoiseandinterferenceinthecommunicationchannel,leadingtoalossinsignalquality.Therefore,QAMblinddetectionisanimportanttasktoimprovetheperformanceofcommunicationsystems. HopfieldneuralnetworksarewidelyusedinQAMblinddetectionduetotheirabilitytosolveoptimizationproblems.However,traditionalHopfieldneuralnetworkscanonlyhandlesingle-valuedsignals.Inthispaper,weproposeanovelapproachbasedoncomplex-valuedHopfieldneuralnetworksforQAMblinddetection. 2.LiteratureReview VariousapproachestoQAMblinddetectionhavebeenproposedovertheyears.Theseincludemaximumlikelihoodestimation,decision-directedestimation,andblindchannelequalization.However,theseapproachessufferfromlimitationssuchashighcomputationalcomplexityortheneedforatrainingsequence. HopfieldneuralnetworkshavebeenshowntobeeffectiveinQAMblinddetection.Thesenetworksuseanenergyfunctiontominimizetheerrorbetweenthereceivedsignalandtheoutputofthenetwork.However,traditionalHopfieldneuralnetworksonlyhandlesingle-valuedsignals. Toaddressthisissue,complex-valuedHopfieldneuralnetworkshavebeenproposed.Thesenetworkscanhandlebothrealandimaginarypartsofasignal,andthuscanhandlecomplex-valuedsignals.However,thesenetworkshavenotbeenextensivelystudiedforQAMblinddetection. 3.ProposedMethod WeproposeanovelapproachforQAMblinddetectionbasedoncomplex-valuedHopfieldneuralnetworks.TheproposedapproachusesamultivaluedHopfieldneuralnetworkthattakesintoaccounttherealandimaginarypartsofthereceivedsignal.Thenetworkisdesignedtominimizetheerrorbetweenthereceivedsignalandtheoutputofthenetwork. Theenergyfunctionofthenetworkisgivenby: E(x)=-1/2∑i,jwi,jxixj-∑ibixi+C wherexiisthebinaryvalueofthei-thneuron,wi,jistheweightbetweenthei-thandj-thneurons,biisthebiasofthei-thneuron,andCisaconstant.Theweightsandbi