Federated Learning NIDS

Network intrusion detection systems trained with federated learning, so that IoT networks can detect emerging threats together without sharing their traffic.

Description

Intrusion detection models need traffic from many networks to recognize new attacks, but that traffic is sensitive and rarely shared. With federated learning each network trains the model locally and only model updates are combined.

The group studies federated and incremental network intrusion detection for IoT, where the system also has to adapt to emerging threats over time. See also Adaptive intrusion detection for IoT and IoMT.

People

Antonio PescapèFull Professor · Principal Investigator
Francesco CerasuoloPostdoc
Domenico CiuonzoAssociate Professor

Publications

Journal R. Carillo, F. Cerasuolo, G. Bovenzi, D. Ciuonzo, A. Pescapè, “A Federated and Incremental Network Intrusion Detection System for IoT Emerging Threats”, IEEE Transactions on Network and Service Management, 2026.

← All research topics