联合授权和非授权频带上的认知网络资源分配
首发时间:2022-02-24
摘要:本文针对下行链路认知无线电(Cognitive Radio, CR)网络,提出一种强化学习(Reinforcement Learning, RL)资源分配框架,协调分配授权频段和非授权频段的可访问频谱资源,实现授权频段和非授权频段上认知用户(Secondary Users, SUs)之间的和谐共存。针对不同的业务,建立多个切片,在保证业务服务质量(Quality of Service, QoS)指标的前提下,尽可能提高非授权频段上认知网络的频谱效率(Spectral Efficiency, SE)并降低占用授权频谱资源的成本。将资源分配视为一个约束学习问题,提出一种基于Actor-Critic的切片资源分配方案,在CR网络中建立资源分配和算法之间的映射关系,根据收发器对(Transceiver Pairs, TPs)的信道占用情况输出SUs最优的信道选择和发射功率。结果表明该方案是有效且可行的。
For information in English, please click here
Cognitive Network Resource Allocation Over Jointly Licensed and Unlicensed Bands
Abstract:Aiming at downlink cognitive radio (CR) network, this paper proposes a reinforcement learning (RL) resource allocation framework to coordinate the allocation of accessible spectrum resources in both licensed and unlicensed bands, so as to realize the harmonious coexistence between the secondary users (SUs) in licensed and unlicensed bands. Multiple slices are established for different services. While ensuring the quality of service (QoS) index, the spectral efficiency (SE) of cognitive network in unlicensed band is improved as much as possible and the cost of occupying licensed spectrum resources is reduced. Considering resource allocation as a constrained learning problem, a slice resource allocation scheme based on Actor-Critic is proposed. The mapping relationship between resource allocation and algorithm is established in CR network. According to the channel occupation of transceiver pairs (TPs), the proposed scheme can output the optimal channel selection and transmission power of SUs. The results show that the scheme is effective and feasible.
Keywords: Network Slicing Resource Allocation Reinforcement Learning
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
联合授权和非授权频带上的认知网络资源分配
评论
全部评论0/1000