Dynamic Defense Framework: A Unified Approach for Intrusion Detection and Mitigation in SDN

Various challenges have hindered achieving strong cybersecurity within the dynamic network configurations of Software Defined Networking (SDN). Traditional cybersecurity measures, especially in programmable and dynamic network infrastructures like SDNs, are not sufficient in mitigating cyber threats. The proposed Dynamic Defense method includes preprocessing of data, extracting features, filtering features, detecting the attacks and mitigating them. Furthermore, a novel hybrid Coot-Lyrebird optimization algorithm is developed to specifically choose the most impactful features. The selected features are given to the proposed hybrid network that combines Convolutional Neural Network (CNN), SE-ResNeXt, and Long Short-Term Memory (LSTM) networks. Finally, the proposed Deep Q-Network (DQN) model performs attack mitigation measures. The results indicate that the proposed Dynamic Defense has accuracy of 0.999571%.