DDosTC: A Transformer-Based Network Attack Detection Hybrid Mechanism in SDN.

Area of Science:

• Computer Science
• Network Security
• Artificial Intelligence

Background:

• Software-defined networking (SDN) offers simplified network management but introduces vulnerabilities.
• Centralized control in SDN makes it susceptible to distributed denial-of-service (DDoS) attacks, which can cause network collapse.
• Existing DDoS detection methods in SDN require enhancement for improved efficacy.

Purpose of the Study:

• To design and evaluate a novel hybrid neural network, DDosTC, for detecting DDoS attacks in SDN environments.
• To leverage the strengths of transformers and convolutional neural networks (CNNs) for robust attack detection.
• To validate the performance of DDosTC against state-of-the-art deep learning algorithms using a recent dataset.

Main Methods:

• Development of a hybrid neural network architecture (DDosTC) integrating transformers and CNNs.
• Training and testing the DDosTC model on the CICDDoS2019 dataset, a comprehensive collection of DDoS attack traffic.
• Comparative analysis of DDosTC against leading deep learning-based DDoS intrusion detection algorithms.

Main Results:

• The DDosTC model achieved an average Area Under the Curve (AUC) 2.52% higher than the current optimal model.
• DDosTC demonstrated superior performance in terms of average accuracy, average recall, and F1 score compared to existing methods.
• Experimental results confirm the effectiveness and scalability of the proposed DDosTC structure for DDoS attack detection in SDN.

Conclusions:

• The DDosTC hybrid neural network offers a significant advancement in detecting DDoS attacks within software-defined networks.
• The integration of transformers and CNNs provides a powerful and efficient solution for network security.
• DDosTC represents a promising approach for enhancing the resilience of SDN infrastructures against sophisticated cyber threats.