Enhancing DDoS Defense in the Internet of Energy through Intelligent Task Scheduling Framework

Li Y., Wei M., Nie J., ERCİŞLİ S., Fang K., Gadekallu T. R.

IEEE Network, 2025 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1109/mnet.2025.3589717
  • Dergi Adı: IEEE Network
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Aerospace Database, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Communication Abstracts, Compendex, Computer & Applied Sciences, INSPEC, Metadex, Civil Engineering Abstracts
  • Anahtar Kelimeler: GNN, Information security, Intelligent power grid, Task scheduling
  • Atatürk Üniversitesi Adresli: Evet

Özet

With the advancement of the Internet of Energy (IoE), a vast array of distributed energy resources, storage systems, electric vehicles, and intelligent end-use devices have been integrated into the network. This integration has made the energy system more open and complex. While this heightened level of interconnectivity enhances operational flexibility and systemic efficiency, it simultaneously engenders a host of security vulnerabilities that warrant serious consideration. This paper presents a novel Heterogeneous Graph Neural Network-Based Task Scheduling framework (HGT) for optimizing multi-node task scheduling in fog computing environments. The proposed model leverages graph neural networks (GNN) and reinforcement learning to dynamically allocate computational resources across distributed IoE devices. By integrating Proximal Policy Optimization (PPO), the model continuously refines scheduling policies, enhancing adaptability to dynamic workloads and varying resource availability. Experimental results demonstrate that the HGT model outperforms traditional heuristic and rule-based scheduling approaches, achieving higher task completion rates and improved computational resource utilization. The framework effectively balances computational loads across fog nodes, reducing resource contention and enhancing system reliability.