基于EMDCLEAN的图像去噪方法摘要：魍车闹斩松璞附诘闶侗鸱椒ㄎ醇扑憬诘悴舛戎副辏导致其节点识别效率和识别率低、识别速度慢。为此，基于边缘计算技术设计了新的终端设备远程部署节点智能识别方法。首先采用希尔伯特变换技术，将处于时域的终端设备信号分解为复频域信号的频率和相位，并确定节点关联程度，建立终端设备网络空间模型;然后利用边缘计算技术设计节点参考框架，计算节点子图和最大邻居连通密度测度，从而实现对终端设备远程部署节点的有效识别。实验结果表明：本方法识别终端设备远程部署节点所需的时间更短，且对节点的变化更为敏感，具有较高的识别效率和节点识别率。关键词：边缘计算技术;终端设备;远程部署节点;节点识别;节点关联程度;连通密度：TP399文献标识码：AResearchonIntelligentIdentificationofRemoteDeploymentNodeofYerminalEquipmentBasedonEdgeComputingTechnologyLUYuhua，GAOWenjun，ZHANGWei（Bei激ngZhixinSemiconductorTechnologyCo.，Ltd.，Bei激ng102200，China）Abstract：Thetraditionalnoderecognitionmethodofterminalequipmentdoesnotcalculatethenodemeasureindex，whichleadstothelownoderecognitionefficiencyandrecognitionrateandslowrecognitionspeed.Forthisreason，anewintelligentidentifica第1页共2页tionmethodforremotedeploymentnodesofterminalequipmentisdesignedbasedonedgecomputingtechnology.Firstly，thesignalinthetimedomainisdecomposedintothefrequencyandphaseofthesignalinthecomplexfrequencydomainbyusingtheHilberttransformtechnology，andthenodecorrelationdegreeisdeterminedtoestablishthenetworkspacemodeloftheterminalequipment.Then，thenodereferenceframeisdesignedbyusingedgecomputingtechnology，andthenodesubplotandthemaximumneighborconnectivitydensitymeasurearecalculated，soastorealizetheeffectiveidentificationoftheterminalequipmentremotedeploymentnode.Theexperimentalresultsshowthattheproposedmethodtakeslesstimetoidentifytheremotedeploymentnodesofterminalequipment，andismoresensitivetothechangesofnodes，andhashigherrecognitionefficiencyandnoderecognitionrate.第2页共2页