Fault Detection for Markov Jump Systems Against Deception Attacks Under Critical-Information Protection Mechanism

This paper focuses on the finite-time fault detection (FD) problem for a class of mode information hiding Markov jump systems (MJSs) with deception attacks. To counteract the influence of attacks, a critical-information protection mechanism (CIPM) is designed, where an event generation function is introduced to determine the importance of data. Utilizing the available measurement information, the transmission-case-depended FD filter is designed by the aid of the hidden Markov model (HMM). By constructing the Lyapunov function that includes the information of system mode and transmission situation, sufficient criteria are derived to ensure that the FD system is stochastically finite-time bounded with satisfactory <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$H_{\infty}$</tex> performance. Ultimately, the superiority of the proposed FD scheme is verified through a comparative simulation involving three cases.