电信科学 ›› 2020, Vol. 36 ›› Issue (2): 101-108.doi: 10.11959/j.issn.1000-0801.2020024

• 运营技术广角 • 上一篇    下一篇

以用户为中心的智能化网络运营服务方法

李凯1,刘伟1,罗贵阳2,李静林2   

  1. 1 中国电信股份有限公司北京研究院,北京 102209
    2 北京邮电大学,北京 100876
  • 修回日期:2020-02-07 出版日期:2020-02-20 发布日期:2020-05-19
  • 作者简介:李凯(1976- ),男,博士,中国电信股份有限公司北京研究院高级工程师,主要研究方向为通信网络能力开放、智能运营等|刘伟(1980- ),男,现就职于中国电信股份有限公司北京研究院,主要研究方向为通信网与互联网融合业务,目前专注于智能化运营研究工作|罗贵阳(1992- ),男,北京邮电大学博士生,主要研究方向为融合网络服务计算与协同计算|李静林(1975- ),男,博士,北京邮电大学副教授,主要研究方向为融合网络协同计算与群智计算
  • 基金资助:
    国家自然科学基金资助项目(61472338)

Customer-centered mobile network intelligent operation approach

Kai LI1,Wei LIU1,Guiyang LUO2,Jinglin LI2   

  1. 1 China Telecom Beijing Research Institute,Beijing 102209,China
    2 Beijing University of Posts and Telecommunications,Beijing 100876,China
  • Revised:2020-02-07 Online:2020-02-20 Published:2020-05-19
  • Supported by:
    The National Natural Science Foundation of China(61472338)

摘要:

多种新型业务的迅猛发展,对移动通信系统的用户个性化业务保障能力提出了更高的要求。传统的智能管道运营仅考虑网络当前的流量状态、承载绑定数量等信息,并未考虑到无线用户的使用需求和特点。同时,由于无线网络的指标多、指标间相关性复杂,且不同用户需求对无线网络指标有不同要求,使用基于规则的方式已经无法应对。因此提出一种以用户为中心的智能化网络运营服务方法,通过学习用户QoS保障需求与无线网络指标态势的相关性,构建基于用户需求与无线网络状态的联合服务保障架构,同时还提出了一种基于机器学习与深度学习的智能化网络服务保障方法,并在真实数据上验证了其有效性。

关键词: 智能化网络运营, 智能管道, 资源分配控制

Abstract:

The rapid development of new types of services places greater demands on support ability of the personalized service in mobile communication systems.Traditional intelligent pipe carriers only consider limited factors such as current traffic status of the network and number of loading binding quantity,indicating that they don’t take account of the demands and features of wireless customers.Meanwhile,the rule-based method is overwhelmed due to numerous metrics of the wireless network,the complicated correlation among them and heterogeneous requirements of metrics from various users.Therefore,a user-centric intelligent network operation service was proposed,which utilized machine learning and deep learning method.The service had the ability to learn the correlation between user QoS guarantee requirements and wireless network metric status to construct a joint support architecture based on user requirements and wireless network status.The validity of the method on real data was verified.

Key words: intelligent operation, intelligent pipe, resource allocation control

中图分类号: 

No Suggested Reading articles found!