移动网络中基于人工智能的端到端网络切片技术研究  

Research on end-to-end network slicing technology based on AI in mobile networks

作　　者：李静[1] 董秋丽 叶宇凯 LI Jing;DONG Qiuli;YE Yukai(China Unicom Network Technology Research Institute,Beijing 100048,China;Beijing University of Posts and Telecommunications,Beijing 100876,China)

机构地区：[1]中国联通网络技术研究院,北京100048 [2]北京邮电大学,北京100876

出　　处：《信息通信技术与政策》2020年第10期78-83,共6页Information and Communications Technology and Policy

基　　金：国家自然科学基金项目(No.61971060)资助。

摘　　要：为了应对未来移动网络差异化的垂直行业应用与海量数据分发挑战,利用网络切片技术变革现有网络架构受到广泛认同。围绕移动网络空中接口部分的切片化面临的问题,设计了包括无线接入网虚拟化的移动网络端到端的网络切片架构,提出了基于网络大数据的服务等级协议参数映射方法,讨论了智能网络切片资源管理可行方法,为移动网络中基于人工智能的端到端网络切片实现提供实践参考。In order to cope with the challenges of vertical industry application and massive data distribution of mobile network differentiation in the future,the use of network slicing technology to transform the existing network architecture is widely recognized.Focused on the problems of air interface in mobile network,this paper proposes an end-to-end network slicing architecture including the virtualization of wireless access network,and a mapping method of service level protocol parameters based on network big data.A feasible method of slicing resource management provides practical reference for the realization of AI-based end-to-end network slicing in mobile networks.

关 键 词：网络切片 人工智能 虚拟化 

分 类 号：TN929.5[电子电信—通信与信息系统] TP18[电子电信—信息与通信工程]