Physics-Attested Federated Learning: Securing Collaborative Anomaly Detection in Critical Water Infrastructure
Federated anomaly detection for water utilities gains a physics-based poisoning defence, addressing a gap where adversaries craft plausible gradients that violate real-world process constraints.
Summary written by editorial AI · Source link below
arXiv:2609.34804v2 Announce Type: replace Abstract: Federated learning enables industrial operators to train shared intrusion detection models without disclosing proprietary operational telemetry. However, existing defences operate strictly in update space, leaving aggregators blind to data poisoning; model updates derived from fabricated telemetry remain indistinguishable from honest contributions. We repurpose process invariants, physical relations that always hold in normal operation, such a
Editorial Analysis
Critical infrastructure operators using federated ML for intrusion detection remain vulnerable to poisoning attacks that pass standard gradient checks; physics-aware defences could materially raise the bar for OT adversaries.
Assess whether your OT anomaly-detection models incorporate domain-specific plausibility checks beyond statistical gradient norms.
Forward-looking interpretation drafted by editorial AI under human review — not a reproduction of the source. See methodology.
External link — opens at arXiv Crypto & Security in a new tab.
More from the OT/IoT Security Desk
- Turning Up the Heat: Inside the Attacks on Europe’s Refrigeration System3d
- CANMentor: A Large-Small Model Collaboration Framework for CAN Bus Intrusion Detection3d
- AMORT-V1: Checkpointed Post-Quantum Attestation for Industrial Controllers3d
- FBI, CISA warn of cyberattacks targeting refrigeration systems4d
- Harden cooling systems now, feds warn critical infrastructure operators amid attacks4d