Impulsive stabilizing control driven by historical DoS data

In this article, we propose a new impulsive control strategy which stabilizes linear systems under denial-of-service (DoS) attacks. Unlike most existing results, our design enables the controller to learn control parameters from historical DoS data, with no prior information (e.g., duration and frequency parameters, periodicity or probabilistic distribution) of the attacker required at the design stage. We prove that our design is effective as long as the DoS attack does not persist indefinitely. We also provide a numerical example to illustrate the robustness against unexpected DoS intensification which renders an existing design ineffective. Furthermore, a biological system is provided to show practical applications of the designed controller.