The Graph Neural Network with Wasserstein Generative Adversarial Network for botnet detection in smart city IoT
Mahesh Kumar Thota, P M Prathibhavani, K R Venugopal · 2024
One of the main risks to the security and stability of Internet of Things (IoT) networks in smart cities is botnet assaults. The identification of complex and emerging botnet attacks has not been entirely contained by traditional methods of botnet detection. In this study, we present a new hybrid approach, called GNN-WGAN, to efficiently detect bots in IoT-based smart city networks by integrating Graph Neural Network and Wasserstein Generative Adversarial Network. The suggested method, which ultimately aims to improve the precision and robustness against botnet detection in dynamic IoT networks, effectively uses WGANs to generate synthetic botnet traffic patterns to add more training data and GNNs to capture dependencies between and interactions across network topology. The experimental results demonstrate the effectiveness of the GNN-WGAN technique in accurately recognizing botnet activities, which enhances the security and resilience of IoT networks in smart cities.