电信科学 ›› 2019, Vol. 35 ›› Issue (7): 60-68.doi: 10.11959/j.issn.1000-0801.2019176

• 专题:5G • 上一篇    下一篇

面向5G雾计算中基于Q-learning的安全中继节点选择方法

涂山山1,2(),于金亮1,2,孟远1,2,WWAQAS M3,刘雷4   

  1. 1 北京工业大学信息学部,北京100124
    2 可信计算北京市重点实验室,北京100124
    3 清华大学电子工程系,北京 100084
    4 北京机电工程研究所,北京 100074
  • 修回日期:2019-07-01 出版日期:2019-07-20 发布日期:2019-07-22
  • 作者简介:涂山山(1983- ),男,北京工业大学信息学部、可信计算北京市重点实验室副教授、硕士生导师,主要研究方向为云计算、MEC、信息安全技术。|于金亮(1996- ),男,北京工业大学信息学部、可信计算北京市重点实验室硕士生,主要研究方向为雾计算、机器学习、信息安全技术。|孟远(1995- ),男,北京工业大学信息学部、可信计算北京市重点实验室硕士生,主要研究方向为雾计算、强化学习、信息安全技术。|WAQAS M(1985- ),男,清华大学电子工程系博士生,主要研究方向为 5G 网络、MEC、信息安全技术。|刘雷(1982- ),男,北京机电工程研究所高级工程师,主要研究方向为航空电子、机电、机器人和智能系统。
  • 基金资助:
    国家自然科学基金资助项目(61801008);国家重点研发计划经费资助项目(2018YFB0803600);北京市自然科学基金资助项目(L172049);北京市教委科研计划经费资助项目(KM201910005025)

Secure relay node selection method based on Q-learning for fog computing in 5G network

Shanshan TU1,2(),Jinliang YU1,2,Yuan MENG1,2,M WWAQAS3,Lei LIU4   

  1. 1 Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China
    2 Beijing Key Laboratory of Trusted Computing, Beijing 100124, China
    3 Department of Electronics and Engineering, Tsinghua University, Beijing 100084, China
    4 Beijing Electro-Mechanical Engineering Institute, Beijing 100074, China
  • Revised:2019-07-01 Online:2019-07-20 Published:2019-07-22
  • Supported by:
    The National Natural Science Foundation of China(61801008);The National Key Research and Development Program of China(2018YFB0803600);Beijing Natural Science Foundation(L172049);The Scientic Research Common Program of Beijing Municipal Commission of Education(KM201910005025)

摘要:

提出了一种基于 Q-learning 的最优双中继节点选择方法。首先构建了基于社会意识的安全雾计算结构模型,然后在该模型下设计了基于 Q-learning 算法的最优双中继节点选择方法,实现了在动态环境下对最优双中继节点的选择,最后对密钥生成速率、双中继节点选择速度和动态环境中双中继节点的选择准确率进行了分析。实验结果表明,该方案能有效地在动态环境中选择最优双中继节点,算法迅速收敛达到稳定,最优中继节点选择速度得到有效提升。

关键词: Q-learning, 雾计算, 5G网络, 社会意识, 物理层安全

Abstract:

A Q-learning-based optimal dual-relay node selection method was proposed. Firstly, a security fog computing structure model based on social awareness was constructed, and then an optimal dual-relay node selection method based on Q-learning algorithm was designed under this model, which achieved the selection of optimal dual-relay nodes in dynamic environment. Finally, the key generation rate, the selection speed of dual-relay nodes and the selection accuracy of dual-relay nodes in dynamic environment were analyzed. The experimental results show that the scheme can effectively select the optimal dual-relay nodes in dynamic environment, the algorithm converges rapidly to a stable level, and the selection speed of the optimal relay node is effectively improved.

Key words: Q-learning, fog computing, 5G network, social awareness, physical layer security

中图分类号: 

No Suggested Reading articles found!