欢迎来到天天文库
浏览记录
ID:25776861
大小:452.00 KB
页数:16页
时间:2018-11-22
《基于lms算法的自适应组合滤波器中英文翻译【完》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、CombinedAdaptiveFilterwithLMS-BasedAlgorithms´Abstract:Acombinedadaptivefilterisproposed.ItconsistsofparallelLMS-basedadaptiveFIRfiltersandanalgorithmforchoosingthebetteramongthem.Asacriterionforcomparisonoftheconsideredalgorithmsintheproposedfilter,wetaketheratiob
2、etweenbiasandvarianceoftheweightingcoefficients.Simulationsresultsconfirmtheadvantagesoftheproposedadaptivefilter.Keywords:Adaptivefilter,LMSalgorithm,Combinedalgorithm,Biasandvariancetrade-off1.IntroductionAdaptivefiltershavebeenappliedinsignalprocessingandcontrol,a
3、swellasinmanypracticalproblems,[1,2].Performanceofanadaptivefilterdependsmainlyonthealgorithmusedforupdatingthefilterweightingcoefficients.ThemostcommonlyusedadaptivesystemsarethosebasedontheLeastMeanSquare(LMS)adaptivealgorithmanditsmodifications(LMS-basedalgorithm
4、s).TheLMSissimpleforimplementationandrobustinanumberofapplications[1–3].However,sinceitdoesnotalwaysconvergeinanacceptablemanner,therehavebeenmanyattemptstoimproveitsperformancebytheappropriatemodifications:signalgorithm(SA)[8],geometricmeanLMS(GLMS)[5],variables
5、tep-sizeLMS(VSLMS)[6,7].EachoftheLMS-basedalgorithmshasatleastoneparameterthatshouldbedefinedpriortotheadaptationprocedure(stepforLMSandSA;stepandsmoothingcoefficientsforGLMS;variousparametersaffectingthestepforVSLMS).Theseparameterscruciallyinfluencethefilteroutput
6、duringtwoadaptationphases:transientandsteadystate.Choiceoftheseparametersismostlybasedonsomekindoftrade-offbetweenthequalityofalgorithmperformanceinthementionedadaptationphases.WeproposeapossibleapproachfortheLMS-basedadaptivefilterperformanceimprovement.Namely,w
7、emakeacombinationofseveralLMS-basedFIRfilterswithdifferentparameters,andprovidethecriterionforchoosingthemostsuitablealgorithmfordifferentadaptationphases.ThismethodmaybeappliedtoalltheLMS-basedalgorithms,althoughwehereconsideronlyseveralofthem.Thepaperisorganize
8、dasfollows.AnoverviewoftheconsideredLMS-basedalgorithmsisgiveninSection2.Section3proposesthecriterionforevaluationandcombinationofadaptivealgorithms.Simulationresults
此文档下载收益归作者所有