简介:TheadvantageofATMtechnologyforstatisticallymultiplexingnetworkresourcesbydifferentusersinvokesthecompetitionofnetworkresources.Thiscompetitionhassomedamagingeffectsonmessagetransmitting.TheresearchfortestingmethodsandtechnologyofATMisimportantforustodevelopATMnetworks,topromotetheirconstruction,toensurethatvariousATMdevicesprovidedbydifferentprovidersinteroperateproperlyandtoprovideexcellentservicesfortelecommunicationusers.Afterdiscussingtheabstracttestingmethodofconformancetestingandtheabstractdescriptionofprotocoltesting,thethesisputsforwardakindofabstracttestingconfigurationforATMtestingandakindofabstractdescriptionmethodfortestingcase.Fromtheangleofapplication,thethesisdiscussesthebasicrulesforATMtesting.Afterthat,thethesispointsoutthattheATMtestingmustbemadefromthefouraspectsofnormaltesting,conformancetesting,performancetestingandinteroperabilitytesting.Thegeneraltestingmethods,generalconfigurationandconnectionforATMtestingandtheselectionofATMtestingitemsarediscussed,respectively,inthefouraspects.CombinedwiththecharacteristicsofthetrafficssupportedbyATMsystems,thethesisdiscussesseveralkindsoftrafficmodelsforconductingATMtesting.Onthebasisofstudyingvarioustrafficmodels,theauthorclassifiestrafficmodelsforATMtestingasthreekinds:periodictrafficmodels,stochastictrafficmodelsandmanualtrafficmodels.Theparametersofdescribingtrafficmodelsandtheircalculatingmethodarediscussedinthethesis.Anewkindofperiodiccell-sequencetrafficmodelisproposedinthisthesis.Theperiodiccell-sequencetrafficmodelhasexcellentlinearlydescendingcharacteristicsinthecellinterval.ThetrafficmodelsdiscussedinthisthesiscanbeappliedtovariousATMtestingcases.Fromthepointofnormaltesting,physicallayertesting,ATMlayertesting,AALlayertesting,networkmanagementtesting
简介:Inthispaper,abroadbandmultimediaapplicationplatform,BaMAP,isintroduced.TheBaMAPaimstooffercontentandserviceprovidersagoodintegratedenvironmentforefficientcontentproduction,management,deliveryandsharing.Itsupportslarge-scalehierarchicalstorage,aswellasmechanismsforcontentdistributionandqueriesfromremotenodes.Theon-demandservice,partoftheBaMAP,offerstwoaccessinterfaces,DVB-CandIP,whichallowtheplatformtorunonmostoftheexistingaccessnetworks.
简介:Theapplicationofcellularneuralnetworks(CNN)forsolvingpartialdifferentialequations(PDEs)isinvestigatedinthispaper.TwokindsofthePDEs,theheat-conductionequationandPoisson'sdquation,areconsideredtobetypicalexamples.TheycanbecomputedinrealtimebyusingtheCNN,whiletheCNN'shardwareisimplementedbytheintegratedOP-AMP.Theexperimentalresultsshowthatthehardwareperformenceisinagreementwiththatgivenbythecomputersimulation.Therefore,theCNNisanewpowerfultoolforsolvingPDEs.
简介:这篇文章处理导致一个混合linear/non-linear模型评价问题的调遣目标追踪的问题。为调遣追踪系统,扩大了Kalman过滤器(EKF)或粒子过滤器(PF)传统地被用来估计状态。在这篇文章,排斥了粒子过滤器(MPF)处于一个混合linear/non-linear模型评价问题为申请被介绍。MPF是Kalman过滤器(KF)和PF的联合。它因此认为两个他们有利并且能被用于混合linear/non-linear基础,在有条件地线性的状态用KF被估计,非线性的状态用PF被估计的地方。模拟结果证明MPF保证评价精确性并且在调遣追踪申请的目标与PF和EKF相比减轻潜在的计算负担问题。