简介:针对网络态势感知中的预测精度问题,提出了基于广义径向基函数(RBF)神经网络的网络安全态势预测方法。该方法利用K-means聚类算法确定RBF的数据中心和扩展函数,并采用最小均方算法调整权值,得出态势值前后之间的非线性映射关系,并进行态势预测。仿真试验表明,该方法能较准确获得态势预测结果,提高网络安全的主动安全防护。
简介:TheDiracsymbolisusedtorepresentthediscretecomplexHopfieldneuralnetworkmodel.Thesignal-to-noisetheoryandthecomputernumericalsolu-tionaremadetoanalysethestoragecapacityofthemodel.Thestoragecapacityra-tioofthemodelequalstothatoftheHopfieldmodel.Finally,usingthemodeltorecognizethe4-levelgreyorcolorpatternsisdiscussed.
简介:Asthetheoryofthefractionalorderdifferentialequationbecomesmaturegradually,thefractionalorderneuralnetworksbecomeanewhotspot.TherobuststabilityofaclassoffractionalorderHopfieldneuralnetworkwiththeCaputoderivativeisinvestigatedinthispaper.ThesufficientconditionstoguaranteetherobuststabilityofthefractionalorderHopfieldneuralnetworksarederivedbymakinguseofthepropertyoftheMittag-Lefflerfunction,comparisontheoremforthefractionalordersystem,andmethodoftheLaplaceintegraltransform.Furthermore,anumericalsimulationexampleisgiventoillustratethecorrectnessandeffectivenessofourresults.
简介:ThemaingoalofroutingsolutionsistosatisfytherequirementsoftheQualityofService(QoS)foreveryadmittedconnectionaswellastoachieveaglobalefficiencyinresourceutilization.InthispaperproposesasolutionbasedonHopfieldneuralnetwork(HNN)todealwithoneofrepresentativeroutingproblemsinuni-castrouting,i.e.themulti-constrained(MC)routingproblem.ComputersimulationshowsthatwecanobtaintheoptimalpathveryrapidlywithournewLyapunovenergyfunctions.