Summary
Detecting unusual behavior in industrial control systems is important to prevent attacks, but ensuring these detection tools work well under pressure is challenging. The authors show that the ability of these systems to handle attacks is directly related to how robust their machine learning components are against adversaries. They created a way to measure resilience by considering how disturbances are absorbed and how recovery happens in a network of detectors. Their tests on a water system revealed surprising effects, like making one component stronger can sometimes lower overall system strength. This work helps guide how to design and certify safer industrial control systems.
Industrial control systemsAnomaly detectionAdversarial robustnessResilience theoryCyber-physical systemsMachine learningAbsorption capacityRecovery trajectoryOperational technologySecurity certification
Abstract
Anomaly-based intrusion detection systems in industrial control systems (ICS) and operational technology (OT) environments are increasingly required to meet formal resilience criteria: absorbed adversarial disturbances, graceful degradation under sustained attack, and certified system-level guarantees. Existing resilience frameworks for cyber-physical systems define absorb-recover-adapt trajectories at the architectural level but do not treat machine learning anomaly detectors as first-class components, leaving a gap between component-level robustness evaluation and system-level resilience certification. In this paper, we establish that adversarial robustness in ICS anomaly detection is a specific instantiation of system resilience, and formalise this connection by mapping four resilience constructs, i.e. disturbance class, absorption capacity, recovery trajectory, and degradation function, onto the adversarial machine learning setting. We derive a compositional resilience bound for heterogeneous ICS detection networks, showing that the binding constraint on system-level resilience is the coupling-adjusted absorption capacity of each node along the attack path, not the per-node capacity -- so the binding node need not be the weakest one. Empirical validation on the BATADAL water distribution system benchmark demonstrates that the resulting metrics surface operationally significant phenomena invisible to standard benchmarks: the absorption-degradation divergence under adversarial training, and the paradox that hardening the binding node in isolation reduces system-level resilience. Implications for ICS architecture design and certification standards are discussed.