Transformer-Based Intrusion Detection for Securing Medical Applications in 5G IoMT Networks

Paper ID: 37

Abstract

This study introduces a transformer-based Network Intrusion Detection System (NIDS) for real-time anomaly detection in 5G IoMT networks. Trained on the WUSTL-HDRL-2024 dataset, the model excels in detecting cyberattacks like DDoS, MiTM, and ransomware, achieving state-of-the-art AUC and F-scores. Leveraging attention mechanisms for interpretability, it addresses class imbalance and enhances cybersecurity for connected medical devices.

Keywords