CNN-LSTM Model for Mitigation of DDoS Attacks in Software-Defined Networks

Abstract

The Internet, as the world's largest computer network, has evolved beyond a mere repository of information to become an indispensable tool driving modern society. Its dynamic nature enables communication, interaction, and access to diverse information sources, shaping how we navigate the digital landscape. However, this ubiquity also makes it vulnerable to Distributed Denial of Service (DDoS) attacks, where malicious actors deliberately disrupt system resources. As network operators transition to Software Defined Networks (SDN) for improved visibility and control, the centralized architecture of SDN becomes susceptible to DDoS attacks, compromising network integrity. In this study, we developed a novel approach for mitigating DDoS attacks by integrating a CNN-LSTM model, achieving promising results with a loss of 0.0131, Accuracy of 0.9952, Precision score of 0.9918, recall score of 0.9961, and F1_score of 0.9939. This fusion of CNN-LSTM models into SDN-based DDoS mitigation systems holds significant potential for enhancing network security and resilience.