| 摘要: |
| 密集异构网络(Dense Heterogeneous Network,DHN)通过部署小基站可以提升网络容量和用户速率，但小基站的密集部署会产生巨大的能耗和严重的干扰，进而影响系统的能量效率(Energy Efficiency,EE)和频谱效率(Spectral Efficiency,SE)。在保证用户服务质量(Quality of Service,QoS)需求的前提下，为了联合优化系统的能量效率和频谱效率，研究了密集异构网络中下行链路的资源分配(Resource Allocation,RA)问题。首先，将频谱和小基站发射功率分配问题建模为联合优化系统能量效率和频谱效率的多目标优化问题；其次，提出了基于单策略多目标强化学习(Single-strategy Multi-objective Reinforcement Learning,SMRL)的资源分配算法求解所建立的多目标优化问题。仿真结果表明，与基于单目标强化学习的资源分配算法相比，所提算法可以实现系统能量效率和频谱效率的联合优化，与基于群体智能算法的资源分配算法相比，所提算法的系统能量效率提高了1%~1.5%，频谱效率提高了1.3%~2.5%。 |
| 关键词: 密集异构网络（DHN） 资源分配 强化学习 多目标优化 |
| DOI：10.20079/j.issn.1001-893x.211220001 |
|
| 基金项目:国家自然科学基金青年基金项目(61901043)；北京信息科技大学“勤信人才”培育计划项目(QXTCP B202101) |
|
| Resource allocation strategy based on multi-objective optimization in dense heterogeneous network |
| JIN Donghui,CHEN Shuo,WANG Zhangang |
| (School of Information and Communication Engineering,Beijing Information Science and Technology University,Beijing 100101,China) |
| Abstract: |
| Dense heterogeneous network（DHN） increases network capacity and user rates by deploying small base stations,but dense deployment of small base stations can generate significant energy consumption and serious interference,which in turn can affect the energy efficiency(EE) and spectral efficiency(SE) of the system.Under the premise of ensuring the quality of service(QoS) requirements of users,in order to jointly optimize the EE and SE of the system,the resource allocation problem of downlink in DHN is studied.Firstly,the spectrum and small base station transmitting power allocation problem is modeled as a multi-objective optimization problem to jointly optimize system EE and SE.Secondly,the problem of multi-objective optimization is solved by the resource allocation algorithm based on single-strategy multi-objective reinforcement learning(SMRL).The simulation results show that compared with the resource allocation algorithm based on single-objective reinforcement learning,the proposed algorithm can achieve the joint optimization of EE and SE of the system,and compared with the resource allocation algorithm based on swarm intelligence algorithm,the system EE of the proposed algorithm is increased by 1%~1.5%,and the SE is increased by 1.3%~2.5%. |
| Key words: dense heterogeneous network(DHN) resource allocation reinforcement learning multi-objective optimization |
