简介：MPLS(Multi-ProtocolLabelSwitching)VPN(VirtualPrivateNetwork)traffichasbeendeployedwidely,butcurrentlyonlysupportsunicast.ThispaperbrieflyintroducesseveralavailableMPLSVPNmulticastapproaches,andthenanalyzestheirdisadvantages.Anovelmechanismthatusestwo-layerlabelstacktosupportMPLSVPNexplicitmulticastisproposedandtheprocessisdiscussedindetail.Thescalabilityandperformanceoftheproposedmechanismarestudiedanalytically.Theresultshowsthatoursolutionhasgreatadvantageoverthecurrentlyavailableschemeintermsofsavingcorenetworkbandwidthandimprovingthescalability.

简介：Averyinexpensiverecive-onlysatellitereceiverhighbandwidthtrafficfromafeeder.Thereforetheconnectionbetweenthefeederandthereceiverisuni-directional.Theexistingroutingprotocols,suchasLink-StateandDistance-Vector,aredesignedonthepremisethatanylinksarebidirectionalandtheycannothandletheuni-directionallinks.Inthispaper,adynamicmulticastroutingprotocolisproposed,whichcanhandleuni-directionalnetworks,Thisprotocolcanalsoadapttothedynamicchangeofthetopolgoyofthenetworkandhasgoodscalability.AformaldescriptionoftheprotocolbyPetrinetisgiven.Liveness,deadlockfreeandotherpropertiesoftheprotocolarepreoved.

简介：Inthispaper,basedonant-algorithm,weproposeaQoSmulticastroutingschemeinInternet.Wefirstdescribeant-algorithmmodelandgiveant-networkmodel,thenpresentanapproachusingant-algorithmtooptimizethemulticastrouteswithQoSconstaints.Finally,simulationshasbeenmadetoshowtheefficiencyoftheapproachintheenvironmentofOPNETsimulationsoftware,andthesimulationresultsshowthattheproposedapproachcanfindthebestoptimalmulticastrouteswhichcansatisfythedelay-boundedrequirementandavoidcongestednodesassoonaspossible.

简介：在里面无线多点传送，网络编码最近作为实质的改进吸引了注意到包重发计划。然而，发现最小化重发的最佳的网络代码的问题是难的解决或接近。这份报纸论述二个计划减少重发的数字为可靠高效地多点传送。一个人是用网络基于改进Vandermonde矩阵(VRNC)编码的重发，其它是用网络基于适应改进Vandermonde矩阵(AVRNC)编码的重发。用VRNC计划，发送者都选择包接收装置与改进Vandermonde失去了并且编码他们矩阵；当接收装置收到足够编码的重发包时，所有失去的包能被恢复。与AVRNC计划，发送者能在每重发包，然后改进Vandermonde矩阵发送以后从所有接收装置获得恢复信息能被更新，因此减少编码并且译码的复杂性。我们的建议计划能完成假定重发包的理论更低的界限无损失，并且就重发包损失而言接近理论更低的界限。模拟结果证明建议算法能高效地减少重发的数字，因此改善的传播效率。

简介：在大规模上在因特网上由于部署因特网协议(IP)的困难多点传送，覆盖多点传送被看作了一种有希望的选择在最近的年里开发multicast通讯。然而，存在覆盖多点传送解决方案受不了高费用在提交树的multicast维持节点的州的信息。一个无状态的覆盖多点传送计划被建议，在哪个multicast路由信息被一个花蕾过滤器(BF)编码并且没有对维持提交树的multicast的任何需要，包含了进包头。我们的计划利用节点异质，在物理拓扑学并且层次地的最近信息由把几何坐标分到所有覆盖节点构造运输树桩覆盖拓扑学。更重要地，计划使用BF技术识别提交树的multicast的节点和连接，它改进提交的效率并且减少假积极的提交环。分析并且模拟结果证明这个建议能完成高提交效率和好可伸缩性。

简介：ThewavelengthassignmentwithmultiplemulticastrequestsinfixedroutingWDMnetworkisstudied.Anewmulticastdynamicwavelengthassignmentalgorithmispresentedbasedonmatchingdegree.First,thewavelengthmatchingdegreebetweenavailablewavelengthsandmulticastroutingtreesisintroducedintothealgorithm.Then,thewavelengthassignmentistranslatedintothemaximumweightmatchinginbipartitegraph,andthismatchingproblemissolvedbyusinganextendedKuhn-Munkresalgorithm.Thesimulationresultsprovethattheoveralloptimalwavelengthassignmentschemeisobtainedinpolynomialtime.Atthesametime,theproposedalgorithmcanreducetheconnectingblockingprobabilityandimprovethesystemresourceutilization.

简介：Multicastcapability,includingmulticastaddressandmulticastroutingmechanisms,atthenetworklayerisnecessaryinordertoreducethebandwidthrequirementsofmultiparty,multicastapplications.Basedonhierarchicalautonomousstructureinaccordancewiththeself-organizationtopologiesofInternet,thepaperputsforwardamulticastaddressmanagementschemethatisshowntoberobustandscalable.Connectioncontrolhierarchy(CCH)basedonmaster/slaverelationshipandasimpleefficientbuildingalgorithmofmulti-pointconnectionisalsobuilt.Thepaperalsodescribesthenormaloperationsofmulticastaddressmanagementandmulti-pointconnectioncontroller.Throughsimulationexperiment,HAMCMandDDMofMulticastAddressAllocationarecompared.HAMintegratesthemeritsofCMandDDM,whichisefficientasawhole,robustandscalable.CCHraisestheefficiencyofconnectioncontrol,andishighlyrobust,flexibleandscalable.

简介：这篇文章在直角的频率部门multiplexing(OFDM)与min率要求限制处理multicast资源分配问题系统。由于为非线性、组合的优化的禁止地高的复杂性，原来的问题被放松并且形成一个标准优化问题的reformulated。根据Karush-Kuhn-Tucker(KKT)由理论推导调节，二项提议为optimality表现为必要标准。而且，包括副载波赋值和力量分配，一个二拍子的圆舞资源分配计划在实际实现的提议的一个基础上被建议。多点传送，min率基于组顺序，副载波以一种贪婪方式被分配最大化能力。当副载波赋值是坚定的时，建议力量分配能完成最佳的性能因为min率与可接受的复杂性抑制了能力最大化。模拟结果显示建议计划接近到彻底的搜索与可以忽略的能力差距获得的最佳的资源分配，并且更加超过相等的力量分发。同时，多点传送对在OFDM系统的资源利用显著地有益。

简介：这篇文章学习多限制最少费用多点传送在在稠密的波长部门multiplexing(IP/DWDM)上的因特网协议的路由问题联网。处理这个问题，individual-difference-based量基因算法(IDQGA)被建议。这个算法由介绍一个适应旋转角度步决心计划和基于组的量变化操作在染色体之中考虑单个差别。模拟在网络拓扑学上被进行。结果显示与另外的启发式的算法相比，IDQGA在解决服务(QoS)的质量上有更好最佳的性能多点传送在IP/DWDM网络的路由问题并且被强壮的坚韧性，高成功比率和优秀能力在全球寻找上描绘。

简介：Thedissertationstudiesthearchitectureofbroadbandintelligentnetwork(BIN)andimplementationofvideoconferencebasedonATMnetworkandofmulticastroutingproblemonvideoconferenceusingBINundertheaidoftheimportanttask-StudyofINandB-ISDNIntegration(69896244)-sponsoredbyNationalNaturalScienceFund.Alotofpointsofviewandsolutionsarepresent.Themaincontributionsinthedissertationareasfollows:(1)DesignthenewarchitectureofBINandimprovethearchitectureofBINproposedbyITU-T.ItiseasytosetuptheconnectionsofvideoconferenceusingthenewarchitectureofBIN.(2)PresentthedetailedschemethatBINcontrolsandimplementsvideoconferencewithoutspecializedresourcefunction,studytheschemeofallthemediaonvideoconferencetransmittedandswitchedunderthecontrolofBIN,anddiscusshowtoimplementpoint-to-multipointcommunicationsusingBIN.(3)FirstpresenttheschemeofimplementingmulticastroutingalgorithmsonvideoconferenceusingBIN.(4)Explicitlyintroduceaseriesofconceptsincludingneighboringnode,neighboringnodeset,neighboringedge,thenproposeanAdjustableDynamicMulticast(ADM)routingalgorithmwhichissuitableforvideoconference,analyzeitsperformances,andprovethealgorithmisappropriateandfeasible.(5)Developadegree-constraineddynamicmulticast(DADM)routingalgorithmwhichcanfindlesscostmulticastroutingtree.TheresultsofsimulationandexperimentsconfirmthatDADMalgorithmcanfindthelesscostmulticasttree.Finally,analyzetherelationbetweenthesizeofdegree-constrainedandvalueofmulticasttreecost.(6)Presentadelay-constraineddynamicmulticastroutingalgorithmsuitableforvideoconferencecharacteristic,andadviseaseriesofschemesreducingthedelayoftransmittingallthemediaonvideoconference.(7)FirstproposeimplementingthereliablemulticastroutingtransportprotocolusingBIN,therebyACKimplosionproblemcanbesolved.Theprotocolonlyrequiresfin

简介：Thestablemulticastflowaggregation(MFA)problemininternetprotocol(IP)overopticalnetworkunderthedynamicalscenarioisstudied.Givenanopticalnetworktopology,thereisasetofheadendsandaccessroutersattachedtotheopticalnetwork,inwhicheachheadendcanprovideasetofprograms(IPmulticastingflows)andeachaccessrouterrequestsasetofprograms,wefindasetofstablelight-treestoaccommodatetheoptimallyaggregatedmulticastIPflowsiftherequestsofaccessrouterschangeddynamically.Weintroduceaprogramcorrelationmatrixtodescribethepreferenceofendusers'requests.AstheoriginalMFAproblemisNP-complete,aheuristicapproach,namedmostcorrelatedprogramfirst(MCPF),ispresentedandcomparedwiththeextendedleasttreefirst(ELTF)algorithmwhichistopology-aware.SimulationresultsshowthatMCPFcanachievebetterperformancethanELTFintermsofstabilitywithnegligibleincrementofnetworkresourceusage.

简介：新multilayered内部卫星高度高度平台(幸运)系统路由算法被建议，它主要基于multipath路由保证网络reliability.The建议multipath路由计划主要依靠切开在不同路径之间的交通使多重routes.Linear编程的利用是的最好成为为multipathselection.The使用的主要方法主要限制到服务(QoS)的质量(延期和连接利用)被考虑了到我

简介：AMinimizingIntermediateMulticastRoutingprotocol(MIMR)isproposedfordynamicmulti-hopadhocnetworks.InMIMR,multicastsessionsarecreatedandreleasedonlybysourcenodes.Ineachmulticastsessionprocess,thesourcenodekeepsalistofintermediatenodesanddestinations,whichisencapsulatedintothepacketheaderwhenthesourcenodesendsamulticastpacket.Nodesreceivingmulticastpacketsdecidetoacceptorforwardthepacketaccordingtothelist.Dependingontopologymatrixmaintainedbyunicastrouting,theshortestvirtualhierarchyroutingtreeiscon-structedbyimprovedDijkstraalgorithm.MIMRcanachievetheminimumnumberofintermediatenodes,whicharecomputedthroughthetree.Nocontrolpacketistransmittedintheprocessofmul-ticastsession.Loadofthenetworkislargelydecreased.ExperimentalresultshowsthatMIMRisflexibleandrobustfordynamicadhocnetworks.

简介：The3-stageClosnetworkC(n,m,r)isconsideredasthemostbasicandpopularmultistageinterconnectionnetworkwhichhasbeenwidelyemployedfordatacommunicationsandparallelcomputingsystems.Quitealotofeortshasbeenputontheresearchofthe3-stageClosnetwork.Unfortunately,verylittleisknownforthemultiratemulticastClosnetworkwhichisthemostcomplicatedcase.Firstlyasucientconditionfor1-ratemulticastnetworkstobeSNBisgiven,fromwhicharesultfor2-ratemulticastnetworkstobeWSNBcaneasilybegotten.Furthermore,byusingareservation-schemerouting,morespecificresultfor2-ratemulticastnetworkstobeWSNBcanbeobtainedforthecaseofoneofthemexceeding1/2.

简介：Inthispaper,anAntColonySystem(AC)basedheuristicalgorithmispresentedtofindthemulti-constraineddynamicQualityofService(QoS)multicastrouting.WealsoexplorethescalabilityoftheACalgorithmandmulticasttreebyusing"Pull"modeinsteadof"Push"mode,andtheimprovementonthetimecomplexityofACalgorithmbyusinganewdatastructure,I.e.,apointerarrayinsteadoftheprevious"matrix"structure.Ourextensivetestsshowthatthepresentedalgorithmcanfindtheglobaloptimumorsuboptimum,andhasagoodscalabilitywithdynamicadaptationtothechangeofmulticastgroup,andgivesbetterperformanceintermsofthetotalcostthanothertwoalgorithms.

简介：Weexperimentallydemonstratethemultiplesignalmodulationonasingleclass10Gverticalcavitysurfaceemittinglaser(VCSEL)carrierat1310nmfornextgenerationmulticast-enableddatacenternetworks.A10Gbit/sdatasignalisdirectlymodulatedontoasinglemodeVCSELcarrier.Tomaximizecarrierspectralefficiency,a2GHzreferencefrequency(RF)clocktoneissimultaneouslymodulatedontheVCSELphaseattribute.TheinherentVCSELorthogonalpolarizationbistabilitywithchangingbiascurrentisfurtherexploitedintransmissionofapolarizationbasedpulsepersecond(PPS)timingclocksignal.Therefore,wesimultaneouslytransmita10Gbit/sdirectlymodulateddata,2GHzphasemodulatedRFandapolarization-basedPPSclocksignalsusingasinglemode10GHzbandwidthVCSELcarrier.Itisthefirsttimethatasingleclass10GVCSELcarrierisreportedtotransmitadirectlymodulateddata,phasemodulatedRFclockandpolarizationbasedPPStimingsignalsimultaneouslyinasinglewavelength.AofG.652singlemodefibre(SMF)transmissionover3.21kmisexperimentallyattained.Areceiversensitivityof-15.60dBmisexperimentallyobtainedforthedirectlymodulated10Gbit/sdatasignal.A3.21-km-longSMFtransmissionintroducesapenaltyof0.23dBtothedatasignal.Thecontributionofa2GHzphasemodulatedRFandapolarization-basedPPSclocksignaltothispenaltyisfoundtobe0.03dB.AnRFsingle-sideband(SSB)phasenoisevaluesof-82.36dBc/Hzand-77.97dBc/Hzareattainedwithoutandwithsimultaneousdirectlymodulateddataandpolarization-basedPPSclocksignalsrespectivelyfora3.21-km-longSMFtransmission.Thisworkprovidesanalternativeefficientandcosteffectivetechniqueforsimultaneoushigh-speedmultipleinformationtransmissiontodifferentnetworknodeswithinadatacenternetworkthroughsharednetworkinfrastructure.

简介：PIM-SM(ProtocolIndependentMulticast-SparseMode)isamainmulticastroutingpro-tocolintheIPv6(InternetProtocolversion6).Itcanuseeitherasharedtreeorashortestpathtreetodeliverdatapackets,consequentlythemulticastIPlookupenginerequires,insomecases,twosearchestogetacorrectlookupresultaccordingtoitsmulticastforwardingrule,anditmayresultinanewrequirementofdoublingthelookupspeedofthelookupengine.TheordinarymethodtosatisfythisrequirementinTCAM(TernaryContentAddressableMemory)basedlookupenginesistoexploitparallelismamongmultipleTCAMs.However,traditionalparallelmethodsalwaysinducemorere-sourcesandhigherdesigndifficulty.Weproposeinthispaperanovelapproachtosolvethisproblem.ByarrangingmulticastforwardingtableinclasssequenceinTCAMandmakingfulluseoftheintrinsiccharacteristicoftheTCAM,ourapproachcangettherightlookupresultwithjustonesearchandasingleTCAM,whilekeepingthehardwareoflookupengineunchanged.ExperimentalresultshaveshownthattheapproachmakeitpossibletosatisfyforwardingIPv6multicastpacketsatthefulllinkrateof20Gb/swithjustoneTCAMwiththecurrentTCAMchip.

简介：

简介：在这份报纸，我们比较了联合网络隧道编码的表演(JNCC)为多点传送当独占时，用低密度同等值支票(LDPC)的继电器网络作为隧道代码编码，Convolutional编码或(XOR)编码的网络在中间的继电器节点使用了。多点传送继电器传播是二个固定继电器节点在第二在作出贡献的传播计划的一种类型在基础收发器车站(BTS)和一双活动车站之间的端对端的传播跳跃。我们认为一个方法和二个方法多点传送评估位错误率(BER)和产量性能的情形。是否使用XOR网络在中间的继电器节点编码，被看了那，一样的传播因此在更少的时间槽变得可能产量性能能被改进。而且我们也在建议系统模型，差异和multiplexing获得在被考虑了讨论了二种可能的情形。它值得通知那BER和产量为LDPC代码完成了比对讨论的所有计划的Convolutional代码好。