基于强化学习的异构无线网络资源管理算法  

Heterogeneous Wireless Network Resource Management Algorithm Based on Reinforcement Learning

作　　者：冯陈伟[1] 袁江南[1] 

机构地区：[1]厦门理工学院光电与通信工程学院,厦门361024

出　　处：《电信科学》2015年第8期99-106,共8页Telecommunications Science

基　　金：福建省中青年教师教育科研项目A类资助项目(No.JA14233);国家自然科学基金青年科学基金资助项目(No.61202013);福建省自然科学基金资助项目(No.2015J01670)~~

摘　　要：为了充分利用各种无线网络的资源,需要实现异构网络的融合,而异构网络的融合又面临接入控制与资源分配的问题。为此,提出一种基于强化学习的异构无线网络资源管理算法,该算法引入D2D(device-to-device)通信模式,并可以根据终端不同的业务类型、终端移动性及网络负载条件等状态,选择合适的网络接入方式。同时,为降低存储需求,采用神经网络技术解决连续状态空间问题。仿真结果表明,该算法具有高效的在线学习能力,能够有效地提升网络的频谱效用,降低阻塞率,从而实现自主的无线资源管理。In order to make full use of the resources of all kinds of wireless network, the integration of heterogeneous network is necessary. However, when it comes to the heterogeneous network integration, the problems of call request access control and resource management emerge. A reinforcement-learning-based algorithm was presented for heterogeneous wireless network resource management. D2D （device-to-device） communication was introduced into the proposed algorithm and the appropriate network for access could be selected according to different traffic types, terminal mobility, network load status and so on. Meanwhile, to reduce the storage requirement, the neural network technology was used to solve the problem of continuous state space. Simulation results show that the proposed algorithm has an efficient learning ability to achieve autonomous radio resource management, which effectively improves the spectrum utility and reduces the blocking probability.

关 键 词：异构无线网络 接入控制 资源管理 强化学习 Q学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]