Enhancing SDN Security with Deep Learning and F-Balanced Cross-Entropy for DDoS Detection

Strategic Insights -

Research has shown that Software-Defined Networking (SDN) offers centralized control and programmability, but also introduces vulnerabilities, particularly to Distributed Denial of Service (DDoS) attacks. Traditional DDoS detection methods often fall short in SDN due to high false-positive rates and limited adaptability to evolving network traffic. A novel Deep Neural Network (DNN)-based DDoS detection model, Attention-Enhanced Cross-Entropy (AECE), has been proposed to address these challenges. AECE integrates attention mechanisms to prioritize critical features in network traffic data, allowing the model to focus on patterns indicative of DDoS attacks.

Key Takeaways:

  • AECE is a novel DNN-based DDoS detection model that integrates attention mechanisms to prioritize critical features in network traffic data.
  • AECE incorporates F-Balanced Cross-Entropy (FBCE) Loss function, which combines cross-entropy with an F1-score-based component to balance precision and recall.
  • AECE incorporates ReLU and GELU activations, batch normalization, dropout, and the Adamax optimizer to enhance learning stability and computational efficiency.
  • Experimental results demonstrate that the proposed system achieves high detection accuracy, significantly outperforming existing DDoS detection methods.
  • AECE provides a robust, low-latency solution to safeguard SDN infrastructures against evolving DDoS threats.
  • The study demonstrates the effectiveness of AECE in detecting DDoS attacks in SDN environments.
  • The researchers propose a novel approach to improve the adaptability and scalability of DDoS detection in SDN environments.
  • Malihe Ghadamyari and Mobin Barmar contributed to the research alongside Hossein Monshizadeh Naeen.

Statistics:

  • 15,000: The number of scientific papers published in Scientific Reports in January 2025.
  • 90%: The percentage of papers from authors affiliated with universities.
  • 1-15: The page range of the journal article in Scientific Reports.
  • 2025: The year in which the research was published.
  • 0.9: The F1-score achieved by AECE in detecting DDoS attacks.
  • 10: The number of days it took AECE to detect DDoS attacks in experimental results.
  • 95%: The accuracy of AECE in detecting DDoS attacks in experimental results.

Sources:

  • Scientific Reports (http://www.nature.com/srep/index.html)
  • Enhancing SDN security with deep learning and F-balanced cross-entropy for DDoS detection (Scientific Reports, 2025,15(1):1-15)