Call for Paper - Upcoming Issues

Deep Reinforcement Learning for Spectral Efficiency in Terahertz-Enabled 6G RANs

International Journal of Electronics and Communication Engineering
© 2026 by SSRG - IJECE Journal
Volume 13 Issue 3
Year of Publication : 2026
Authors : Sheetal Vishal Deshmukh, Shahanawaj Ahamad, P S V Srinivasa Rao, G. Meena Devi
10.14445/23488549/IJECE-V13I3P109
pdf
How to Cite?

Sheetal Vishal Deshmukh, Shahanawaj Ahamad, P S V Srinivasa Rao, G. Meena Devi, "Deep Reinforcement Learning for Spectral Efficiency in Terahertz-Enabled 6G RANs," SSRG International Journal of Electronics and Communication Engineering, vol. 13,  no. 3, pp. 119-128, 2026. Crossref, https://doi.org/10.14445/23488549/IJECE-V13I3P109

Abstract:

Terahertz communication is a fundamental component towards the realization of the highest possible data rate that is currently being conceived in the 6G Radio Access Networks. The bands of Terahertz frequencies have large path loss and significant molecular absorption and thus require accurate and dynamic resource management. To maximize Spectral Efficiency while balancing energy constraints, this research proposes a robust Deep Reinforcement Learning system that is built upon a Multi-Objective Double Deep Q-network framework, including the environment, sensing, intelligence, and performance layers built using the Deep MIMO ray-tracing dataset to create a high-fidelity digital twin of the Terahertz channel. Experiments show that the framework can overcome overestimation bias by performing extensive normalization, state-vectorizing, and multi-objective reward shaping to stabilize the learning process, which achieves a rapid convergence and higher stability than conventional Deep Q Network methods. The proposed model achieves significant improvements in Spectral Efficiency (24.1 bps/Hz), Energy Efficiency (4.8 Gbits/J), Throughput Satisfaction Rate (up to 95%), and sub framework beam alignment latency (0.8 ms). The results show the effectiveness of the Deep Reinforcement Learning methodologies in solving complex propagation problems in the 6G Terahertz communication systems.

Keywords:

Deep Reinforcement Learning, 6G Wireless Networks, Terahertz Communication, Radio Access Networks, Multi Objective Double Deep Q-Network.

References:

[1] Shen Wang et al., “Explainable AI for 6G Use Cases: Technical Aspects and Research Challenges,” IEEE Open Journal of the Communications Society, vol. 5, pp. 2490-2540, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Zakria Qadir et al., “Towards 6G Internet of Things: Recent Advances, use Cases, and Open Challenges,” ICT Express, vol. 9, no. 3, pp. 296-312, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Bokang Francis et al., “The Terahertz Channel Modeling in Internet of Multimedia Design In-Body Antenna,” International Journal of E-Health and Medical Communications, vol. 13, no. 4, pp. 1-17, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Ruijin Sun et al., “A Comprehensive Survey of Knowledge-Driven Deep Learning for Intelligent Wireless Network Optimization in 6G,” IEEE Communications Surveys and Tutorials, vol. 28, pp. 1099-1135, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Zhiyuan You et al., “Hierarchical Beamforming Optimization for ISAC-Enabled RSU Systems in Complex Urban Environments,” Sensors, vol. 25, no. 21, pp. 1-27, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Nada Elsokkary et al., “Reinforcement Learning and the Metaverse: A Symbiotic Collaboration,” Artificial Intelligence Review, vol. 59, pp. 1-58, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Mehdi Setayesh, and Vincent W. S. Wong, “Viewport Prediction, Bitrate Selection, and Beamforming Design for THz-Enabled 360° Video Streaming,” IEEE Transactions on Wireless Communication, vol. 24, no. 3, pp. 1849-1865, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Mourice O. Ojijo, and Olabisi E. Falowo, “A Survey on Slice Admission Control Strategies and Optimization Schemes in 5G Network,” IEEE Access, vol. 8, pp. 14977-14990, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Zilu Liu, and Qichao Xu, “RIS-Enhanced UAV-Assisted Transmission Rate Optimization with Anti-Jamming,” Physical Communication, vol. 71, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Shereen S. Omar et al., “Capacity Enhancement of Flying-IRS Assisted 6G THz Network Using Deep Reinforcement Learning,” IEEE Access, vol. 11, pp. 101616-101629, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Nonis Wara et al., “Multi-Agent PPO-Based Resource Optimization for Full-Duplex RIS-Aided NOMA-ISAC Systems,” IEEE Open Journal of the Communications Society, vol. 6, pp. 9802-9820, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Jianjun Ma et al., “Terahertz Channels in Atmospheric Conditions: Propagation Characteristics and Security Performance,” Fundamental Research, vol. 5, no. 2, pp. 526-555, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Helder Fontes et al., “Improving ns-3 Emulation Performance for Fast Prototyping of Routing and SDN Protocols: Moving Data Plane Operations to Outside of ns-3,” Simulation Modelling Practice and Theory, vol. 96, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Dimitrios G. Selimis et al., “Path Loss, Angular Spread and Channel Sparsity Modeling for Indoor and Outdoor Environments at the Sub-THz Band,” Physical Communication, vol. 66, pp. 1-9, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Wentao Yu et al., “An Adaptive and Robust Deep Learning Framework for THz Ultra-Massive MIMO Channel Estimation,” IEEE Journal on Selected Topics in Signal Processing, vol. 17, no. 4, pp. 761-776, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Hai-Bo Xu et al., “A Beer-Lambert Law-Based General Acceleration Approach for Numerical Computation of Radiative Heat Transfer within Silica Aerogel,” Thermal Science and Engineering Progress, vol. 67, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Khizra Asaf, Bilal Khan, and Ga-Young Kim, “Wireless Lan Performance Enhancement Using Double Deep Q-Networks,” Applied Sciences, vol. 12, no. 9, pp. 1-20, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Jiangbo Tang et al., “Deep-Reinforcement-Learning–Guided Resource Allocation and Task Offloading for 6G Edge Intelligence,” Computer Communications, vol. 245, pp. 1-17, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Nan Wang, and Blesson Varghese, “Context-Aware Distribution of Fog Applications using Deep Reinforcement Learning,” Journal of Network and Computer Applications, vol. 203, pp. 1-14, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Amjad Iqbal et al., “Twin Delayed Deep Deterministic Policy Gradient-based Physical Layer Security and SEE in RIS-aided UAV Communication,” Computer Networks, vol. 274, pp. 1-14, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Gabriel Pimenta de Freitas Cardoso, Paulo Henrique Portela De Carvalho, and Paulo Roberto de Lira Gondim, “Joint Spectrum Allocation and Power Control for D2D Communication and Sensing in 6G Networks using DRL-Based Hyper-Heuristics Author Links Open Overlay Panel,” Computer Networks, vol. 276, 2026.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Preksha Shah et al., “Energy Efficiency Optimization and DQN-based Power Allocation in UAV-IRS Assisted 6 G System,” ICT Express, pp. 1-8, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Qianhong Cong, and Wenhui Lang, “Double Deep Recurrent Reinforcement Learning for Centralized Dynamic Multichannel Access,” Wireless Communications and Mobile Computing, vol. 2021, pp. 1-10, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Faysal Marzuk, Andres Vejar, and Piotr ChoĊ‚da, “Deep Reinforcement Learning for Energy-Efficient 6G V2X Networks,” Electronics, vol. 14, no. 6, pp. 1-23, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Md. Alamgir Hossain, “Deep Q-Learning Intrusion Detection System (DQ-IDS): A Novel Reinforcement Learning Approach for Adaptive and Self-learning Cybersecurity,” ICT Express, vol. 11, no. 5, pp. 875-880, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Yao Li et al., “Alleviating the Estimation Bias of Deep Deterministic Policy Gradient via Co-Regularization,” Pattern Recognition, vol. 131, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Brice Jibia et al., “A Hybrid Deep Reinforcement Learning and Genetic Algorithm Approach for Optimized Resource Allocation and Reconfigurable Intelligent Surface Deployment in 6G Holographic Communication,” International Journal of Intelligent Networks, vol. 6, pp. 265-282, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Wentao Yu et al., “AI and Deep Learning for Terahertz Ultra-Massive MIMO: From Model-Driven Approaches to Foundation Models,” Engineering, vol. 56, pp. 14-33, 2026.
[CrossRef] [Google Scholar] [Publisher Link]