The faculty member, Assistant Lecturer Shahin Muhammad Ali, has published a dataset titled "Medical IoT Traffic with Cyber Attacks" on the Kaggle platform. This is an advanced synthetic dataset that simulates medical network traffic with high fidelity, supported by open-source code for reproducibility.
The dataset includes data at two levels (Packet-level and Session-level) for 20 medical devices over a period of 7 days, featuring realistic daily operation patterns and network protocol simulations, including the TCP Handshake, in addition to modeling cross-device network congestion.
The data encompasses multi-stage cyberattacks (Recon, Spoof, Exfil, Inject) at an auto-tuned rate of ~15%. It also introduces realistic noise, including packet loss, delay, jitter, and outliers, alongside ambiguous labels to increase model complexity.
The dataset supports research in medical network security, intrusion detection systems (IDS), and machine learning. It provides ready-to-use tools such as temporal splitting and an entropy-based baseline for evaluation, along with automated reports and statistics.
🔗 For viewing and downloading:
https://www.kaggle.com/datasets/shahin121212/medical-iot-traffic-with-cyber-attacks
