通信学报 ›› 2021, Vol. 42 ›› Issue (4): 44-61.doi: 10.11959/j.issn.1000-436x.2021096

所属专题: 边缘计算

• 专题:面向未来移动网络的大规模组网关键技术 • 上一篇    下一篇

大规模无线通信网络移动边缘计算和缓存研究

黄永明1,2, 郑冲1, 张征明1, 尤肖虎1,2   

  1. 1 东南大学信息科学与工程学院,江苏 南京 210096
    2 紫金山实验室,江苏 南京 211111
  • 修回日期:2021-03-31 出版日期:2021-04-25 发布日期:2021-04-01
  • 作者简介:黄永明(1977- ),男,江苏吴江人,博士,东南大学教授、博士生导师,主要研究方向为智能 5G/6G 移动通信、毫米波无线通信等。
    郑冲(1994- ),男,湖北荆州人,东南大学博士生,主要研究方向为智能无线通信、移动边缘计算、边缘智能、物联网、深度强化学习和联邦学习等。
    张征明(1994- ),男,安徽阜阳人,东南大学博士生,主要研究方向为无线大数据、机器学习、5G移动网络、无人机辅助通信和资源管理等。
    尤肖虎(1962- ),男,山东济宁人,博士,东南大学教授、博士生导师,主要研究方向为移动通信系统和信号处理及其应用等。
  • 基金资助:
    国家自然科学基金资助项目(61720106003);国家重点研发计划基金资助项目(2018YFB1800801)

Research on mobile edge computing and caching in massive wireless communication network

Yongming HUANG1,2, Chong ZHENG1, Zhengming ZHANG1, Xiaohu YOU1,2   

  1. 1 School of Information Science and Engineering, Southeast University, Nanjing 210096, China
    2 Purple Mountain Laboratory, Nanjing 211111, China
  • Revised:2021-03-31 Online:2021-04-25 Published:2021-04-01
  • Supported by:
    The National Natural Science Foundation of China(61720106003);The National Key Research and Devel-opment Program of China(2018YFB1800801)

摘要:

面向未来6G移动通信的大规模网络移动边缘计算与缓存技术,首先,介绍了大规模无线网络下移动边缘计算和缓存的架构与原理,并阐释了移动边缘计算和缓存技术在大规模无线网络中的必要性和普适性。接着,从计算卸载、边缘缓存、多维资源分配、用户关联和隐私保护这5个关键问题出发,综述和分析了移动边缘计算和缓存赋能大规模无线网络时会引入的新型关键问题以及对应的解决方案研究,并进一步指出了未来的发展趋势和研究方向。最后,针对隐私保护问题,提出了一种基于联邦学习的隐私保护方案,并通过仿真结果表明所提方案能够同时保护用户数据隐私且改善系统服务质量。

关键词: 大规模无线网络, 移动边缘计算, 缓存, 隐私保护, 联邦学习

Abstract:

For the large-scale network mobile edge computing and caching technology of future 6G mobile communications, firstly, the architectures and principles of mobile edge computing and caching in large-scale wireless networks were introduced, and the necessity and universality were clarified.Then, from the perspective of the five key issues in the mobile edge computing and caching enabled large-scale wireless network, including computing offloading, edge caching, multi-dimensional resource allocation, user association and privacy protection, the recent researches and further pointed out the future development trends and research directions were reviewed and analyzed.Finally, for the privacy preservation issue, a federated learning based privacy-preserving scheme was proposed.Simulation results show that the proposed scheme can simultaneously preserve user privacy and improve the quality of service.

Key words: massive wireless network, mobile edge computing, caching, privacy protection, federated learning

中图分类号: 

No Suggested Reading articles found!