摘要
介绍了无线网络中的强化学习算法,认为由于强化学习算法与环境交互并动态决策的特点,其对复杂网络环境有着较强的适应能力;然后针对无线网络中的强化学习方法的应用场景做了概述,并给出了两个基于强化学习的无线接入技术案例:毫米波技术的切换技术和Multi-RAT接入技术。可以看到:智能的无线接入技术由于具备充分挖掘和扩展无线网络资源的潜力,能够显著提高无线网络用户的体验。
In this paper, the application of reinforcement learning in wirelessnetwork is briefly introduced. Due to the characteristics of interacting withenvironment and dynamic decision making, reinforcement leaning algorithm hasstrong adaptability to complex network environment. Then the application scenariosof reinforcement learning method in wireless network are summarized, and twocases of wireless access technology based on reinforcement learning are given:handoff policy of mm Wave Het Nets and multi-rat access control. Intelligent accesscontrol of wireless network is powerful in exploiting wireless network resources,which can improve the quality of experiences of mobile users.
作者
严牧
孙耀
冯钢
YAN Mu SUN Yao FENG Gang(University of Electronic Science and Technology of China, Chengdu 611731, China)
出处
《中兴通讯技术》
2018年第2期10-14,46,共6页
ZTE Technology Journal
基金
国家自然科学基金(61631005
61471089)
中央高校基本科研基金(ZYGX2015Z005)
关键词
未来无线网络
切换
接入控制
强化学习
future wireless network
handoff
access control
reinforcement learning