基于Bayes评估的5G SA网络切片安全映射算法  

5G SA network slicing security mapping algorithm based on Bayesian evaluation

作　　者：蒋建峰 张运嵩[2] 张娴 张趁香[3] 安淑梅 JIANG Jianfeng;ZHANG Yunsong;ZHANG Xian;ZHANG Chenxiang;AN Shumei(School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210023;Suzhou Industrial Park Institute of Services Outsourcing,Suzhou 215123;School of Computer Science&Technology,Soochow University,Suzhou 215006;Ruijie Networks Co.,Ltd.,Fuzhou 350002)

机构地区：[1]南京邮电大学计算机学院,南京210023 [2]苏州工业园区服务外包职业学院,苏州215123 [3]苏州大学计算机科学与技术学院,苏州215006 [4]锐捷网络股份有限公司,福州350002

出　　处：《南京信息工程大学学报（自然科学版）》2022年第5期527-534,共8页Journal of Nanjing University of Information Science & Technology（Natural Science Edition）

基　　金：国家自然科学基金(61702351);江苏省博士后研究基金(2018K009B);江苏省专业带头人高端研修项目(2020GRFX074);江苏省高校“青蓝工程”项目(202010)。

摘　　要：5G网络的发展促使工业互联网垂直行业战略转型升级.5G SA(Stand⁃Alone)网络将用户面功能(UPF)下沉到业务区,控制面留在大区中心,所有业务数据均通过5G边缘计算接口(MEC)访问业务应用服务器,用户访问自身服务器流量全部在内部完成,符合网络安全需求,业务路径时延最小.通过优化的粒子群算法(PSO),构造基于网络物理节点资源和链路资源的适值函数,并且引入贝叶斯评估方法计算物理节点隔离因子,将所有节点平均隔离因子设为阈值作为约束条件,保证网络切片的隔离性能,通过迭代求解来调整网络切片间的路径资源提升网络性能.经过智慧矿山实际5G网络测试,结果显示算法能够提升收益成本比10%,提高链路利用率19%,提高上行速率20%~50%,充分保障矿区网络的服务质量.The development of 5G network promotes the strategic transformation and upgrading of the vertical in⁃dustry of the industrial Internet.The 5G SA(Stand⁃Alone)network sinks the User Plane Function(UPF)into the business area,and retains the control plane in the center of the large area.All business data is accessed through the 5G Multi⁃access Edge Computing(MEC)interface to the business application server,while all traffic for users to access their own servers is completed internally,which guarantees the network security and minimizes the business path delay.Through the optimized Particle Swarm Optimization(PSO),a fitness function based on network physical node resources and link resources is constructed,in which the Bayesian evaluation is introduced to calculate the physical node isolation factor.The average isolation factor of all nodes is set as the threshold then used as constraint condition to ensure the isolation performance of network slices,and the iterative solution is used to adjust the path resources between network slices to improve network performance.The results of actual 5G network test on smart mine show that the algorithm can increase the revenue⁃cost ratio by 10%,improve the link utilization rate by 19%and the uplink rate by 20%-50%,which fully guarantees the quality of network service in mining areas.

关 键 词：5G独立组网 边缘计算 智慧矿山 下沉UPF 网络软切片 隔离因子 粒子群算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]