Zero Dynamics Attack Detection and Isolation in Cyber-Physical Systems with Event-triggered Communication

TL;DR

The paper tackles Zero Dynamics (ZD) cyber-attacks in cyber-physical systems by introducing an auxiliary system that operates with event-based communication to detect and isolate such attacks. A residual-based scheme is developed, including a self-triggering rule to limit bandwidth while enabling reliable detection, and an augmentation of the plant with a zero-dynamics-free auxiliary state facilitates detection even when attackers have full knowledge of the dynamics. The authors prove stability and show Zeno-freeness via LMIs, and validate the approach through two simulation case studies (quadruple-tank and throttle-to-speed aircraft) illustrating robustness against sophisticated, knowledge-rich adversaries. The work has practical impact for securing CPS against high-resource attackers while reducing communication overhead, with avenues for extending to delayed, nonlinear, and multi-agent settings.

Abstract

This paper investigates the problem of Zero Dynamics (ZD) cyber-attack detection and isolation in Cyber-Physical Systems (CPS). By utilizing the notion of auxiliary systems with event-based communications, we will develop a detection mechanism capable of detecting and isolating the ZD cyber-attack even when the attackers have full knowledge of the dynamics of the auxiliary system and can launch False Data Injection (FDI) attacks on all the communication channels. More specifically, we will utilize a self-triggering rule for the communication channels connecting the auxiliary system with the Command & Control (C&C) center, leveraging its properties to detect the ZD cyber-attack. Finally, the effectiveness and capabilities of our approach are verified and demonstrated through simulation case studies.

Figures (7)

• Figure 1: Overall schematic of the CPS where the input channel is subject to the cyber-attack signal $a_u$ and the auxiliary channel is subject to the cyber-attack signal $\bar{a}_z(t)=a_z(t_j),\forall t \in [t_j,t_{j+1})$.
• Figure 2: The residual $res_z$ when the system is subject to ZD cyber-attack on the input channel and the cyber-attack signal $a_z$ is designed based on (\ref{['eq:CovertAz']}). As demonstrated, the residual will not exceed the threshold.
• Figure 3: The residual (\ref{['eq:Res_t']}) when the system is under the ZD cyber-attack where the residual exceeds the threshold and the cyber-attack is detected.
• Figure 4: The residual $res_z$ when the system is subject to ZD cyber-attack on the input channel and the cyber-attack signal $a_z$ is designed based on (\ref{['eq:CovertAz']}). As demonstrated, the residual will not exceed the threshold.
• Figure 5: The residual $res_z$ when the system is subject to ZD cyber-attack on the input channel and the cyber-attack signal $a_z$ is designed based on (\ref{['eq:CovertAz']}). As demonstrated, the residual will not exceed the threshold.