A time-domain estimation of the ν-gap metric and stability margin with its application to attack defense in cyber-physical systems

To address the limitations of model-based moving target defense (MTD) scheme against stealthy attacks in cyber–physical systems, this paper proposes a novel data-driven MTD (DD-MTD)approach based on the ν-gap metric and the stability margin. To design and implement the traditional model-based MTD in a data-driven framework, the ν-gap metric and the stability margin are estimated in the time domain under the plug-and-play process monitoring and control architecture (PnP-PMCA). Besides, the anomaly detector based on the ν-gap metric aids in distinguishing between faults and stealthy attacks. As an essential part of this study, the stable image representation (SIR) and kernel representation (SKR) are identified using real-time closed-loop data. The proposed approaches are verified via a numerical example and a test on the Mecanum-wheeled mobile robot.