HybridGuard: Enhancing Minority-Class Intrusion Detection in Dew-Enabled Edge-of-Things Networks

The introduction of HybridGuard marks a pivotal development in securing Dew-Enabled Edge-of-Things (EoT) networks, which face increasing threats from sophisticated intrusions. This framework employs a combination of machine learning and deep learning techniques to tackle the critical issue of data imbalance, particularly in detecting minority attack classes. By utilizing mutual information-based feature selection and Wasserstein Conditional Generative Adversarial Networks with Gradient Penalty (WCGAN-GP), HybridGuard enhances detection precision and performance. Its two-phase architecture, DualNetShield, supports advanced traffic analysis and anomaly detection, allowing for more granular threat identification. Evaluated on datasets such as UNSW-NB15, CIC-IDS-2017, and IOTID20, HybridGuard has demonstrated superior performance across diverse attack scenarios, outperforming existing solutions. This advancement is crucial as it not only strengthens the security of EoT networks but also ad…