A Hierarchical Federated Learning-Based Intrusion Detection System for 5G Smart Grids

As the core component of smart grids, advanced metering infrastructure (AMI) provides communication and control functions to implement critical services, which makes its security crucial power companies customers. An intrusion detection system (IDS) can be applied monitor abnormal information trigger an alarm protect AMI security. However, existing models exhibit a low performance are commonly trained on cloud servers, pose major threat user privacy increase delay. To solve these problems, we present transformer-based model (Transformer-IDM) improve detection. In addition, integrate 5G technology into propose hierarchical federated learning (HFed-IDS) collaboratively train Transformer-IDM in networks. Finally, extensive experimental results using real-world dataset demonstrate that proposed approach is superior other approaches terms accuracy cost for IDS.