Paper details stealthy data-driven attacks on closed-loop control systems
Original title: Iterative Behavioral Identification for Covert Misappropriation Attacks
A model-free MITM method identifies plant behavior and tracks arbitrary references while staying under an infinity-norm IDS.
Researchers describe a three-phase framework for covert misappropriation of closed-loop networked control systems. A man-in-the-middle adversary uses only intercepted input-output data to grow a Hankel behavioral model while limiting IDS residuals. A final DeePC-style optimization synthesizes a covert attack. Simulations on a PI-controlled DC motor show full-rank identification and stealthy tracking under noise-free LTI assumptions.
Unconfirmed: Stealth during the online rank-increasing phase is numerically supported, not analytically guaranteed.
- Phase 1 uses a low-amplitude probe inside an analytical stealth bound.
- Hankel rank grows to n + mL in exactly n + mL - 1 steps.
- Stealth is analytical in Phases 1 and 3 and numerical in Phase 2.
- The method assumes noise-free linear time-invariant dynamics.
Impact: Covert reference tracking on a closed-loop control system
From the original article
1 min read at arxiv.orgWe present a data-driven, three-phase framework for covert misappropriation attacks on closed-loop networked control systems. A man-in-the-middle (MITM) adversary, with access only to intercepted input-output data, incrementally identifies the plant's behavioral subspace while maintaining stealth under an infinity-norm intrusion-detection system (IDS).
By Victor Mattos · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.