Call For Paper - Upcoming Conferences
Research Article | Open Access | Download PDF
Volume 13 | Issue 4 | Year 2026 | Article Id. IJECE-V13I4P111 | DOI : https://doi.org/10.14445/23488549/IJECE-V13I4P111

Graph-Enhanced Multi-Agent Deep Reinforcement Learning for Secure Energy-Harvesting Clustering in 6G IoT Sensor Networks

Yuvaraj Duraisamy, I. Poonguzhali, P S V S Sridhar, M. Kavitha

Received Revised Accepted Published
09 Jan 2026 11 Feb 2026 12 Mar 2026 30 Apr 2026

Citation :

Yuvaraj Duraisamy, I. Poonguzhali, P S V S Sridhar, M. Kavitha, "Graph-Enhanced Multi-Agent Deep Reinforcement Learning for Secure Energy-Harvesting Clustering in 6G IoT Sensor Networks," International Journal of Electronics and Communication Engineering, vol. 13, no. 4, pp. 147-155, 2026. Crossref, https://doi.org/10.14445/23488549/IJECE-V13I4P111

Abstract

To meet future 6G ecosystems, the deployment of large-scale Internet of Things (IoT) systems needs to be intelligent with respect to security requirements, intermittent energy supply, and ultra-dense network environments. Conventional clustering and routing methods do not ensure reliability in case the nodes work in energy-harvesting mode and under changing interference. The paper introduces a graph-enhanced multi-agent deep reinforcement learning (GMADRL) model, which will handle secure clustering in the 6G IoT sensor networks. The suggested system has integrated graph neural representations with topological interactions among nodes and multi-agent reinforcement learning in motivating decentralized cluster-head selection, energy-harvesting correction, and threat-informed decision-making. Multi-objective reward maximizes the stability of the clusters, secure communication, the use of energy that is harvested, and the reliability of the routing. The simulation experiments can reveal the ability of the proposed framework to significantly improve the longevity of clusters, their energy balance, adversarial behaviour resistance, and throughput in 6G network conditions. GMADRL has lower packet drop rates, greater energy yield, and balanced convergence behaviour of the agents compared to traditional DRA-based clustering schemes. The results of this study remind us of the capabilities of graph-augmented multi-agent intelligence in providing resilient clustering to future 6G IoT models.

Keywords

6G IoT, Deep Reinforcement Learning, Multi-Agent Systems, Graph Neural Networks, Secure Clustering, Energy Harvesting.

References

  1. Haijun Zhang et al., “Network Slicing based 5G and Future Mobile Networks: Mobility, Resource Management, and Challenges,” IEEE Communications Magazine, vol. 55, no. 8, pp. 138-145, 2017.
    [    CrossRef] [Google Scholar] [Publisher Link]
  2. Q. Hugh, “FPGA-Accelerated Graph Neural Pipelines for Multi-Agent Reinforcement Learning in Dense IoT Mesh Networks,” Journal of Reconfigurable Hardware Architectures and Embedded Systems, vol. 2, no. 3, pp. 1-7, 2025.
    [    Google Scholar]
  3. Rajib Chowdhuri, and Mrinal Kanti Deb Barma, “Node Position Estimation based on Optimal Clustering and Detection of Coverage Hole in Wireless Sensor Networks Using Hybrid Deep Reinforcement Learning,” Research Square, vol. 79, pp. 20845-20877, 2023.
    [    CrossRef] [Google Scholar] [Publisher Link]
  4. Lau W. Cheng, “Spectrum-aware DRL Clustering Protocols for 6G IoT Nodes using Graph Signal Intelligence,” Journal of Wireless Intelligence and Spectrum Engineering, vol. 2, no. 2, pp. 1-7, 2025.
    [    Google Scholar] [Publisher Link]
  5. Yaohua Sun et al., “Application of Machine Learning in Wireless Networks: Key Techniques and Open Issues,” IEEE Communications Surveys & Tutorials, vol. 21, no. 4, pp. 3072-3108, 2019.
    [    CrossRef] [Google Scholar] [Publisher Link]
  6. K. Geetha, “Autonomous IoT Clustering in Ubiquitous Environments using Multi-Agent Graph Reinforcement Learning,” National Journal of Ubiquitous Computing and Intelligent Environments, vol. 2, no. 2, pp. 31-38, 2025.
    [    Google Scholar] [Publisher Link]
  7. Kohyar Bolvary Zadeh Dashtestani, Reza Javidan, and Reza Akbari, “Dynamic Clustering Method for Underwater Wireless Sensor Networks based on Deep Reinforcement Learning,” Journal of Marine Science and Application, vol. 24, pp. 864-875, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  8. F. Rahman, “Secure Routing Mechanisms for Energy-Harvesting IoT using Adversarial-Resilient Multi-Agent Deep Reinforcement Learning,” Transactions on Internet Security, Cloud Services, and Distributed Applications, vol. 2, no. 1, pp. 9-16, 2025.
    [    Google Scholar] [Publisher Link]
  9. W.R. Heinzelman, A. Chandrakasan, and H. Balakrishnan, “Energy-Efficient Communication Protocol for Wireless Microsensor Networks,” Proceedings of the 33rd Annual Hawaii International Conference on System Sciences, Maui, HI, USA, vol. 2, pp. 1-10, 2000.
    [    CrossRef] [Google Scholar] [Publisher Link]
  10. Namrata Mishra, “Scalable Graph-Centric Data Engineering for Energy-Harvesting Intelligence in 6G IoT Networks,” Journal of Scalable Data Engineering and Intelligent Computing, vol. 2, no. 2, pp. 1-8, 2025.
    [    Google Scholar] [Publisher Link]
  11. Hye Yun Kim, “An Energy-Efficient Qos-Aware Clustered Routing Protocol Using Reinforcement Learning for Wireless Sensor Networks,” Journal of Southwest Jiaotong University, vol. 59, no. 1, pp. 161-70, 2024.
    [    CrossRef] [Publisher Link]
  12. Saravanakumar Veerappan, “Robust 6G IoT Communication Models for Assistive Devices Using Graph-Augmented Deep Reinforcement Learning,” Journal of Intelligent Assistive Communication Technologies, vol. 2, no. 1, pp. 57-63, 2025.
    [    Google Scholar] [Publisher Link]
  13. Jun Li et al., “Deploying Graph Neural Networks in Wireless Networks: A Link Stability Viewpoint,” IEEE Wireless Communications Letters, vol. 13, no. 10, pp. 2757-2761, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  14. J. Logeshwaran et al., “Hybrid Optimization for Efficient 6G IoT Traffic Management and Multi-Routing Strategy,” Scientific Reports, vol. 14, pp. 1-17, 2024.
    [    CrossRef] [Google Scholar] [Publisher Link]
  15. Sumit Ramswami Punam, “Automated Distributed Learning Pipelines for Multi-Agent Graph Intelligence in 6G IoT Systems,” SECITS Journal of Scalable Distributed Computing and Pipeline Automation, vol. 2, no. 2, pp. 18-27, 2025.
    [    Google Scholar] [Publisher Link]
  16. Zonghan Wu et al., “A Comprehensive Survey on Graph Neural Networks,” IEEE Transactions on Neural Networks and Learning Systems, vol. 32, no. 1, pp. 4-24, 2021.
    [    CrossRef] [Google Scholar] [Publisher Link]
  17. S. Phani Praveen et al., “Design of an Iterative Method for Adaptive Federated Intrusion Detection for Energy-Constrained Edge-Centric 6G IoT Cyber-Physical Systems,” Scientific Reports, vol. 15, pp. 1-16, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  18. Saravanakumar Veerappan, “Secure Graph-Driven Communication Architectures for Energy-Harvesting 6G IoT Sensor Networks,” Transactions on Secure Communication Networks and Protocol Engineering, vol. 2, no. 1, pp. 42-49, 2025.
    [    Google Scholar] [Publisher Link]
  19. Lucian Busoniu, Robert Babuska, and Bart De Schutter, “A Comprehensive Survey of Multiagent Reinforcement Learning,” IEEE Transactions on Systems, Man, and Cybernetics Part C, vol. 38, no. 2, pp. 156-172, 2008.
    [    CrossRef] [Google Scholar] [Publisher Link]
  20. Zhi Zhou et al., “Edge Intelligence: Paving the Last Mile of Artificial Intelligence with Edge Computing,” Proceedings of the IEEE, vol. 107, no. 8, pp. 1738-1762, 2019.
    [    CrossRef] [Google Scholar] [Publisher Link]
  21. C.N. Vanitha, and P. Anusuya, “Towards Sustainable Wireless Rechargeable Sensor Networks: A Federated Multi-Agent Reinforcement Learning Approach for Cooperative Wireless Charging,” Bulletin of the Polish Academy of Sciences: Technical Sciences, vol. 74, no. 1, pp. 1-12, 2026.
    [    CrossRef] [Google Scholar] [Publisher Link]
  22. Chuhang Wang, Huangshui Hu, and Xinji Fan, “Intelligent Clustering and Routing Protocol for Wireless Sensor Networks using Quantum-Inspired Harris Hawk Optimizer and Deep Reinforcement Learning,” Ad Hoc Networks, pp. 1-12, 2025.
    [    CrossRef] [Google Scholar] [Publisher Link]
  23. P. Joshua Reginald, “Context-Adaptive Intelligent Metasurface Architectures for Reliable Wireless Interaction in Pervasive Environments,” Archives of Electronics, Communication and Emerging Technologies, vol. 1, no. 1, pp. 25-32, 2026.
    [    CrossRef] [Google Scholar] [Publisher Link]
  24. Q. Hugh et al., “An Intelligent Embedded System Architecture for Real-Time Signal Processing in IoT Platforms,” Journal of Integrated VLSI and Signal Intelligence, vol. 1, no. 1, pp. 34-41, 2026.
    [    Google Scholar] [Publisher Link]
  25. N. Arvinth, “Design of Ultra-Low-Power VLSI Architectures for Edge-Intelligent IoT Applications,” National Journal of VLSI Systems and Integrated Circuit Design, vol. 1, no. 1, pp. 1-9, 2025.
    [    Google Scholar] [Publisher Link]