A generalizable dynamic flow pairing method for traffic classification

A generalizable dynamic flow pairing method for traffic classification

ID:39711557

大小:1.17 MB

页数:15页

时间:2019-07-09

A generalizable dynamic flow pairing method for traffic classification_第1页
A generalizable dynamic flow pairing method for traffic classification_第2页
A generalizable dynamic flow pairing method for traffic classification_第3页
A generalizable dynamic flow pairing method for traffic classification_第4页
A generalizable dynamic flow pairing method for traffic classification_第5页
资源描述:

《A generalizable dynamic flow pairing method for traffic classification》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、ComputerNetworks57(2013)2718–2732ContentslistsavailableatSciVerseScienceDirectComputerNetworksjournalhomepage:www.elsevier.com/locate/comnetAgeneralizabledynamicflowpairingmethodfortrafficclassificationJoséCamacho,PabloPadilla,PedroGarcía-Teodoro,JesúsDíaz-Verdejo⇑DepartmentofSignalTheory,Telematic

2、sandCommunication,ResearchCenterofInformationandCommunicationTechnologies,UniversityofGranada,Granada,SpainarticleinfoabstractArticlehistory:ThegoalofnetworktrafficclassificationistoidentifytheprotocolsortypesofprotocolsinReceived11December2012thenetworktraffic.Inparticular,theidentificationofnetwor

3、ktrafficwithhighresourceReceivedinrevisedform13May2013consumption,suchaspeer-to-peer(P2P)traffic,representsagreatconcernforInternetSer-Accepted14June2013viceProviders(ISP)andnetworkmanagers.Mostcurrentflow-basedclassificationAvailableonline22June2013approachesreporthighaccuracywithoutpayingattention

4、tothegeneralizationabilityoftheclassifier.However,withoutthisability,aclassifiermaynotbesuitableforon-lineclas-Keywords:sification.Inthispaper,anumberofexperimentsonrealtraffichelptoelucidatethereasonTrafficclassificationforthislackofgeneralization.ItisalsoshownthatonewaytoattainthegeneralizationPeer-

5、to-peerFlowabilityisbyusingdynamicclassifiers.Fromtheseresults,adynamicclassificationPairingapproachbasedonthepairingofflowsaccordingtoasimilaritycriterionisproposed.Thepairingmethodisnotaclassifierbyitself.Rather,itsgoalistodetermineinafastwaythattwogivenflowsaresimilarenoughtoconcludetheycorrespond

6、tothesamepro-tocol.Combiningthismethodwithaclassifier,mostoftheflowsdonotneedtobeexplic-itlyevaluatedbythelater,sothatthecomputationaloverheadisreducedwithoutasignificantreductioninaccuracy.Inthispaper,asacasestudy,weexplorecomplementingthepairingmethodwithpayloadinspection.Intheexperimentsperforme

7、d,thepairingapproachgeneralizeswelltotrafficobtainedindifferentconditionsandscenariosthanthatusedforcalibration.Moreover,ahighportionofthetrafficunclassifiedbypayloadinspec-tioniscategorizedwiththepairingmethod.Ó2013ElsevierB.V

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。