详细信息
Energy efficiency performance in RIS-based integrated satellite-aerial-terrestrial relay networks with deep reinforcement learning ( SCI-EXPANDED收录 EI收录)
文献类型:期刊文献
英文题名:Energy efficiency performance in RIS-based integrated satellite-aerial-terrestrial relay networks with deep reinforcement learning
作者:Li, Jiao[1];Xue, Huajian[2];Wu, Min[3];Wang, Fucheng[2];Gao, Tieliang[1];Zhou, Feng[4]
第一作者:李敬
通讯作者:Xue, HJ[1];Wu, M[2]
机构:[1]Xinxiang Univ, Key Lab Data Anal & Financial Risk Predict, Xinxiang 453000, Peoples R China;[2]Tongling Univ, Tongling 244000, Peoples R China;[3]Space Engn Univ, Sch Space Informat, Beijing 101400, Peoples R China;[4]Yancheng Inst Technol, Sch Informat Technol, Yancheng 224000, Peoples R China
第一机构:新乡学院
通讯机构:[1]corresponding author), Tongling Univ, Tongling 244000, Peoples R China;[2]corresponding author), Space Engn Univ, Sch Space Informat, Beijing 101400, Peoples R China.
年份:2023
卷号:2023
期号:1
外文期刊名:EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING
收录:;EI(收录号:20234815125075);Scopus(收录号:2-s2.0-85178154123);WOS:【SCI-EXPANDED(收录号:WOS:001114095100001)】;
基金:The authors would like to extend their gratitude to the anonymous reviewers for their valuable and constructive comments, which have largely improved and clarified this paper.
语种:英文
外文关键词:Satellite-aerial-ground integrated networks (SAGINs); Hybrid FSO/RF links; Reconfigurable intelligent surface (RIS); Long short-term memory-deep deterministic policy gradient (LSTM-DDPG); Deep reinforcement learning (DRL)
摘要:Integrated satellite-aerial-terrestrial relay networks (ISATRNs) play a vital role in next-gen networks, particularly those with high-altitude platforms (HAP). This study introduces a new model for hybrid optical/RF-based HAP-enabled ISATRNs, incorporating reconfigurable intelligent surfaces (RIS) on unmanned aerial vehicles (UAVs) to optimize access in dense urban areas. Non-orthogonal multiple access is employed for improved spectrum efficiency. The objective is to jointly optimize UAV trajectory, RIS phase shift, and active transmit beamforming while considering energy consumption. A deep reinforcement learning approach using LSTM-DDQN framework is proposed. Numerical results show the effectiveness of our algorithm over traditional DDQN, with higher single-step exploration reward and evaluation metrics.
参考文献:
正在载入数据...