| 摘要: |
| 为了研究基于深度强化学习（Deep Reinforcement Learning，DRL）的5G异构网络模型的性能，同时在最小化系统能耗并满足不同类型终端用户的服务质量要求的基础上制定合理的资源分配方案，提出了一种基于DRL的近端策略优化算法，并结合一种基于优先级的分配策略，引入了海量机器类型通信、增强移动宽带和超可靠低延迟通信业务。所提算法相较于Greedy和DQN算法，网络延迟分别降低73.19%和47.05%，能耗分别降低9.55%和6.93%，而且可以保证能源消耗和用户延迟之间的良好权衡。 |
| 关键词: 5G 异构网络 资源分配 深度强化学习 |
| DOI：10.20079/j.issn.1001-893x.230215003 |
|
| 基金项目:国家自然科学基金资助项目（61661018）；江苏省基础研究计划青年基金项目（BK20210064）；江苏省双创博士人才项目（JSSCBS20210863）；南京信息工程大学滨江学院科研启动项目（2021r006） |
|
| Priority-based Joint Resource Allocation for Deep Reinforcement Learning in 5G Heterogeneous Networks |
| SONG Duanzheng,GUO Yecai,LI Hui,ZHU Jintao,WANG Hao |
| (1.School of Electronic and Information Engineering,Nanjing University of Information Science & Technology,Nanjing 210044,China;2.Jiangsu Province Engineering Research Center of Integrated Circuit Reliability Technology and Testing System,Wuxi University,Wuxi 214105,China) |
| Abstract: |
| In order to investigate the performance of Deep Reinforcement Learning(DRL)-based heterogeneous network models and develop a reasonable resource allocation scheme based on minimizing system energy consumption and satisfying the quality of service(QoS) requirements of different types of end users,the authors propose a DRL-based Proximal Policy Optimization(PPO) algorithm and introduce massive machine type communication(mMTC),enhanced mobile broadband(eMBB) and ultra ultra-reliable and low latency communications(URLLC) services by combining with a priority-based allocation policy.compared with the Greedy and DQN algorithms,the proposed algorithm reduces the network delay by 73.19% and 47.05%,and the energy consumption by 9.55% and 6.93%,respectively,and can ensure a good trade-off between energy consumption and user latency. |
| Key words: 5G heterogeneous networks resource allocation deep reinforcement learning |
