Instant power allocation for terahertz NOMA communication networks based on Transformer-double deep Q-network

Zhou Yin

doi:10.11959/j.issn.1000-0801.DXKX260150

您当前的位置：

首页 >

文章列表页 >

Instant power allocation for terahertz NOMA communication networks based on Transformer-double deep Q-network

Research and Development | 更新时间：2026-07-21

- Instant power allocation for terahertz NOMA communication networks based on Transformer-double deep Q-network
- Telecommunications Science Vol. 42, Issue 6, Pages: 119-130(2026)
- 作者机构：
  
  中国人民解放军91977部队，北京 100036
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.DXKX260150
  CLC： TN92
- Received：06 March 2026，
  
  Revised：2026-04-11，
  
  Accepted：18 May 2026，
  
  Online First：21 July 2026，
  
  Published：20 June 2026
- 稿件说明：
移动端阅览
周音.基于Transformer-双Q网络的太赫兹NOMA通信网络即时功率分配[J].电信科学,2026,42(06):119-130.

Zhou Yin.Instant power allocation for terahertz NOMA communication networks based on Transformer-double deep Q-network[J].Telecommunications Science,2026,42(06):119-130.
周音.基于Transformer-双Q网络的太赫兹NOMA通信网络即时功率分配[J].电信科学,2026,42(06):119-130. DOI： 10.11959/j.issn.1000-0801.DXKX260150.

Zhou Yin.Instant power allocation for terahertz NOMA communication networks based on Transformer-double deep Q-network[J].Telecommunications Science,2026,42(06):119-130. DOI： 10.11959/j.issn.1000-0801.DXKX260150.

摘要

太赫兹非正交多址接入（non-orthogonal multiple access，NOMA）技术有望成为6G通信系统的关键突破性方案。其核心机制是利用超宽带资源与功率域复用，实现海量用户共享多个子频段，显著提升系统连接容量。要充分释放太赫兹NOMA系统的性能潜力，关键在于要满足服务质量（quality of service，QoS）约束下的快速功率分配优化要求。为此，提出了一种基于Transformer架构的双Q网络模型，通过Transformer学习不同用户分配策略的关联性，并采用双Q网络实现更稳定的决策优化过程。经训练后，该算法可生成适应多种用户分布的即时功率分配策略。实验结果表明，训练完成的模型仅需要毫秒级计算时间，即可实现接近穷举法的高吞吐量性能。该算法展现出较强的实时性与鲁棒性，具有较大工程应用潜力。

Abstract

Terahertz (THz) non-orthogonal multiple access (NOMA) is regarded as a candidate in 6G and beyond systems. By exploiting the ultrabroad bandwidth and power domain

THz-NOMA can realize massive connectivity by assigning each sub-band to different users. To unleash the potential of the THz-NOMA system

it is significant to allocate power quickly under quality of service (QoS) requirements. Focusing on the instant power allocation

a novel Transformer-based double deep Q-network solution adaptive to general user distributions was proposed. Transformer was used to learn the relationships among allocation strategies for different users

and a double deep

Q-network was adopted to achieve a more stable decision optimization process. The simulation results validate that the proposed algorithm realized the throughput close to the optimum given by exhaustive search method within the millisecond level. The proposed method demonstrates high real-time performance and robustness

which suggests its high practicability.

关键词

Keywords

references

唐雄燕 , 李福昌 , 张忠皓 , 等 . 6G网络需求、架构及技术趋势 [J ] . 移动通信 , 2021 , 45 ( 4 ): 37 - 44 .

Tang X Y , Li F C , Zhang Z H , et al . Requirements, architectures and technology trends of 6G network [J ] . Mobile Communications , 2021 , 45 ( 4 ): 37 - 44 .

Wang C X , You X H , Gao X Q , et al . On the road to 6G: visions, requirements, key technologies, and testbeds [J ] . IEEE Communications Surveys & Tutorials , 2023 , 25 ( 2 ): 905 - 974 .

王战将 , 李凯乐 , 张飞翔 , 等 . 面向6G的太赫兹通信技术研究 [J ] . 移动通信 , 2025 , 49 ( 5 ): 121 - 127 .

Wang Z J , Li K L , Zhang F X , et al . A review of terahertz communication technology for 6G [J ] . Mobile Communications , 2025 , 49 ( 5 ): 121 - 127 .

Petrov V , Pyattaev A , Moltchanov D , et al . Terahertz band communications: applications, research challenges, and standardization activities [C ] // Proceedings of the 2016 8th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT) . Piscataway : IEEE Press , 2016 : 183 - 190 .

Akyildiz I F , Han C , Hu Z F , et al . Terahertz band communication: an old problem revisited and research directions for the next decade [J ] . IEEE Transactions on Communications , 2022 , 70 ( 6 ): 4250 - 4285 .

Xue Q , Ji C W , Ma S D , et al . A survey of beam management for mmWave and THz communications towards 6G [J ] . IEEE Communications Surveys & Tutorials , 2024 , 26 ( 3 ): 1520 - 1559 .

Dai L L , Wang B C , Yuan Y F , et al . Non-orthogonal multiple access for 5G: solutions, challenges, opportunities, and future research trends [J ] . IEEE Communications Magazine , 2015 , 53 ( 9 ): 74 - 81 .

Saito Y , Kishiyama Y , Benjebbour A , et al . Non-orthogonal multiple access (NOMA) for cellular future radio access [C ] // Proceedings of the 2013 IEEE 77th Vehicular Technology Conference (VTC Spring) . Piscataway : IEEE Press , 2013 : 1 - 5 .

Ahmed A , Wang X F , Hawbani A , et al . Unveiling the potential of NOMA: a journey to next-generation multiple access [J ] . IEEE Communications Surveys & Tutorials , 2025 , 27 ( 5 ): 3099 - 3164 .

Ding Z G , Adachi F , Poor H V . The application of MIMO to non-orthogonal multiple access [J ] . IEEE Transactions on Wireless Communications , 2016 , 15 ( 1 ): 537 - 552 .

Zhang X F , Han C , Wang X D . Joint beamforming-power-bandwidth allocation in terahertz NOMA networks [C ] // Proceedings of the 2019 16th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON) . Piscataway : IEEE Press , 2019 : 1 - 9 .

Jiao L C , Shao Y L , Sun L , et al . Advanced deep learning models for 6G: overview, opportunities, and challenges [J ] . IEEE Access , 2024 , 12 : 133245 - 133314 .

Mahmood M R , Matin M A , Sarigiannidis P , et al . A comprehensive review on artificial intelligence/machine learning algorithms for empowering the future IoT toward 6G era [J ] . IEEE Access , 2022 , 10 : 87535 - 87562 .

Sanjalawe Y , Fraihat S , Al-E’Mari S , et al . A review of 6G and AI convergence: enhancing communication networks with artificial intelligence [J ] . IEEE Open Journal of the Communications Society , 2025 , 6 : 2308 - 2355 .

Saetan W , Thipchaksurat S . Power allocation for sum rate maximization in 5G NOMA system with imperfect SIC: a deep learning approach [C ] // Proceedings of the 2019 4th International Conference on Information Technology (InCIT) . Piscataway : IEEE Press , 2019 : 195 - 198 .

Xiao L , Li Y D , Dai C H , et al . Reinforcement learning-based NOMA power allocation in the presence of smart jamming [J ] . IEEE Transactions on Vehicular Technology , 2018 , 67 ( 4 ): 3377 - 3389 .

Zhang Y H , Wang X M , Xu Y Y . Energy-efficient resource allocation in uplink NOMA systems with deep reinforcement learning [C ] // Proceedings of the 2019 11th International Conference on Wireless Communications and Signal Processing (WCSP) . Piscataway : IEEE Press , 2019 : 1 - 6 .

Van H H , Guez A , Silver D . Deep reinforcement learning with double Q-learning [C ] // Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16) . Phoenix, Arizona : AAAI Press , 2016 : 2094 - 2100 .

Zhao D B , Liu D R , Lewis F L , et al . Special issue on deep reinforcement learning and adaptive dynamic programming [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2018 , 29 ( 6 ): 2038 - 2041 .

李伟琨 , 姚信威 , 王万良 , 等 . 太赫兹通信中MIMO信道建模与容量分析 [J ] . 计算机工程 , 2015 , 41 ( 4 ): 125 - 128 .

Li W K , Yao X W , Wang W L , et al . MIMO channel modeling and capacity analysis in terahertz communication [J ] . Computer Engineering , 2015 , 41 ( 4 ): 125 - 128 .

Alhaj N A , Jamlos M F , Manap S A , et al . Integration of hybrid networks, AI, ultra massive-MIMO, THz frequency, and FBMC modulation toward 6G requirements: a review [J ] . IEEE Access , 2024 , 12 : 483 - 513 .

Yan L F , Han C , Yuan J H . Energy-efficient dynamic-subarray with fixed true-time-delay design for terahertz wideband hybrid beamforming [J ] . IEEE Journal on Selected Areas in Communications , 2022 , 40 ( 10 ): 2840 - 2854 .

Hu Z F , Han C , Deng Y S , et al . Multi-task deep reinforcement learning for terahertz NOMA resource allocation with hybrid discrete and continuous actions [J ] . IEEE Transactions on Vehicular Technology , 2024 , 73 ( 8 ): 11647 - 11663 .

Han C , Wang Y Q , Li Y B , et al . Terahertz wireless channels: a holistic survey on measurement, modeling, and analysis [J ] . IEEE Communications Surveys & Tutorials , 2022 , 24 ( 3 ): 1670 - 1707 .

Wang J , Wang C X , Huang J , et al . 6G THz propagation channel characteristics and modeling: recent developments and future challenges [J ] . IEEE Communications Magazine , 2024 , 62 ( 2 ): 56 - 62 .

Riazul Islam S M , Avazov N , Dobre O A , et al . Power-domain non-orthogonal multiple access (NOMA) in 5G systems: potentials and challenges [J ] . IEEE Communications Surveys & Tutorials , 2017 , 19 ( 2 ): 721 - 742 .

He C F , Hu Y , Chen Y , et al . Joint power allocation and channel assignment for NOMA with deep reinforcement learning [J ] . IEEE Journal on Selected Areas in Communications , 2019 , 37 ( 10 ): 2200 - 2210 .

Wen M N , Lin R J , Wang H J , et al . Large sequence models for sequential decision-making: a survey [J ] . Frontiers of Computer Science , 2023 , 17 ( 6 ): 176349 .

Yuan W L , Chen J X , Chen S F , et al . Transformer in reinforcement learning for decision-making: a survey [J ] . Frontiers of Information Technology & Electronic Engineering , 2024 , 25 ( 6 ): 763 - 790 .

陈卓 , 冯钢 , 何颖 , 等 . 运营商网络中基于深度强化学习的服务功能链迁移机制 [J ] . 电子与信息学报 , 2020 , 42 ( 9 ): 2173 - 2179 .

Chen Z , Feng G , He Y , et al . Deep reinforcement learning based migration mechanism for service function chain in operator networks [J ] . Journal of Electronics & Information Technology , 2020 , 42 ( 9 ): 2173 - 2179 .

Termehchi A , Bao T N , Syed A , et al . Goal-oriented reinforcement learning in THz-enabled UAV-aided network using supervised learning [J ] . IEEE Open Journal of the Communications Society , 2024 , 5 : 5027 - 5036 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Resource allocation and power optimization for maritime communication of dual USV based on NOMA

A layered collaborative resource allocation algorithm for cell-free RAN aimed at efficient resource utilization

Design and performance analysis of enhanced cognitive NOMA communication system

An optimization algorithm based on deep reinforcement learning for maritime MEC task offloading and resource allocation

SRv6 dynamic performance optimization method based on machine learning

Related Author

Lin Jiongxun

Huang Junkun

Liao Shengtao

WANG Boye

XIA Weiwei

MIAO Weiwei

ZHANG Mingxuan

PAN Yuqing

Related Institution

South China Academy of Advanced Optoelectronics, South China Normal University

School of Electronic Engineering, Guangxi University of Science and Technology

State Grid Suzhou Power Supply Company

Information & Telecommunication Branch, State Grid Jiangsu Electric Power Co., Ltd.

National Mobile Communication Research Laboratory of Southeast University

⁰