AI-Powered DDoS Detection in 6G EH Networks: A New Era in IoT Security

Strategic Insights -

A significant threat to the stability of IoT-enabled infrastructures is emerging due to the rapid expansion of IoT ecosystems, particularly in energy hub (EH) networks. Researchers at the Faculty of Engineering Technology have developed a new AI-powered DDoS detection system, using machine learning models and hybrid ensemble techniques tailored for DDoS detection in IoT-driven 6G EH networks. This system outperforms traditional methods, achieving an accuracy of 99.44% and an F1-score of 66.62% on CICDDOS2019, while maintaining robust performance on other benchmark datasets.

Key Takeaways:

  • The researchers evaluated five machine learning models (Random Forest, Gradient Boosting, Support Vector Machines, Decision Trees, and K-Nearest Neighbors) and their hybrid combinations for DDoS detection in IoT-driven 6G EH networks.
  • The most effective model was RF + KNN, achieving the highest accuracy and F1-score on CICDDOS2019 while maintaining robust performance on other datasets.
  • GB + DT demonstrated superior precision on KDD-CUP, while GB + KNN achieved the highest recall on CICDDOS2019.
  • Smaller training data sizes generally led to performance degradation in F1-score and recall, suggesting potential overfitting with larger datasets.
  • The research underscores the importance of tailoring hybrid ensemble models to the specific properties of datasets and attack types.
  • The system provides a scalable, real-time framework for intrusion detection and enhances the security and resilience of IoT-driven 6G EH networks.

Statistics:

  • The hybrid model RF + KNN achieved an accuracy of 99.44% and an F1-score of 66.62% on CICDDOS2019.
  • GB + DT demonstrated a precision of 70.95% on KDD-CUP.
  • GB + KNN achieved a recall of 66.67% on CICDDOS2019.
  • The use of smaller training data sizes led to a performance degradation in F1-score and recall, with occasional accuracy improvements.

Sources:

  • Discover Applied Sciences, 2025,7(9):1-33
  • Journal of Engineering, September 15, 2025, p 2425
  • https://doi-org.sdpl.idm.oclc.org/10.1007/s42452-025-06716-9 (free journal article)