Improving Model Accuracy in DDoS Prevention by Leveraging GANs

The integration of IoT devices in healthcare enhances patient care and real-time monitoring but also increases vulnerability to Distributed Denial of Service (DDoS) attacks threatening system operations and patient safety. A DDoS attack is one in which the perpetrator seeks to make a system / resource unavailable to its intended users by temporarily or indefinitely disrupting services. We propose an AI-driven, defense strategy for healthcare IoT systems, focusing on anomaly detection through deep learning models. These models detect irregularities in network traffic, distinguishing between routine fluctuations and malicious DDoS attacks. We propose an innovative approach to generate balanced datasets using GANs while improving classifier performance. The novelty of this approach lies in the carefully designed loss function linkage in the GAN training process to address class imbalance and improve DDoS packet classification accuracy. This comprehensive framework combines AI, proactive and threat simulation to protect healthcare IoT systems from evolving cyber threats.