Deep Learning Approach for Optimized DDoS Detection in SDN

---

Résumé

The rise of Software-Defined Network (SDN) offers great flexibility in network infrastructure management. However, this flexibility also introduces critical security challenges, in particular vulnerability to Distributed Denial of Service attacks (DDoS). Traditional detection methods are often ineffective in the face of evolving attack strategies and generate high false positive rates. To solve these problems, this paper proposes a Deep Learning (DL)-based approach for accurate and efficient detection of DDoS attacks in SDN environments. We evaluated the performance of Convolutional Neural Network (CNN), Deep Neural Networks (DNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) and Recurrent Neural Networks (RNN) models for identifying DDoS attacks in SDN networks. The models were trained on a specific SDN dataset. Metrics such as accuracy, precision, recall, and F1-score were used to evaluate the models. The analysis of the results showed that GRU offered the best detection performance with an accuracy rate of 99.45%. These results highlight the potential of DL to improve the resilience of SDN architectures to DDoS.

Mots-clés

Software-Defined Network , Distributed Denial of Service , Deep Learning , models , specific dataset , performance