Contents I. Introduction
Benefiting from the tremendous development of sensing, signal processing, and communication technologies, networked control systems (NCSs) have attracted extensive attention from the areas of system science, control engineering, computer science, robotics, and so on, due to their low cost deployment, easy maintenance, large coverage range, and superb flexibility [1], [2], [3], [4], [5]. One advantage of the NCSs is their remote control capability through networked communication. However, the easy access to the network makes it vulnerable to malicious attackers (see Fig. 1), which worsens their control performance and even causes system divergence or faults. As a representative type of malicious attacks, denial-of-service (DoS) attack could completely paralyze the entire network communication backbone within a certain time period. As pioneer works, Liu et al. [6] concerned with the input-to-state stability of the system under DoS attacks and the impulsive controller. Pan et al. [7] utilized the state errors induced by DoS attacks to design an event-triggered resilient control law to reduce the adverse impact. Zhang et al. [8] addressed an adaptive dynamic programming-based optimal control problem for unknown nonlinear systems suffered by DoS attacks. Ye et al. [9] investigated an observer-based sliding-mode control problem of unmanned marine vehicle systems under DoS attacks. Yang et al. [10] proposed a codesign control scheme composed of fault detection and event-triggered protocol development for a multiarea wind power system under dual DoS attacks. Qi et al. [11] proposed a permissible type-switching strategy between local and networked controllers, whose virtue lies in saving network resources and reducing the impacts of DoS attacks. Kato et al. [12] studied the effect of DoS attacks on linearization methods in stabilization problems and gave an explicit relationship between the initial state and the DoS parameters.
