Event-triggered attack-tolerant tracking control design for networked nonlinear control systems under DoS jamming attacks

Abstract

This paper addresses the problem of event-triggered attack-tolerant tracking control for a networked nonlinear system under spasmodic denial-of-service (DoS) attacks. Compared with some existing results, theexact duration of DoS attacks is not required while only assuming attainable bounds of attack frequency and duration. First, a new event-triggered attack-tolerant fuzzy tracking controller is proposed, to reduce the amount of sensor data transmissions over the sensor-to-controller (S-C) channel while counteracting unknown DoS attacks. Second, for the purpose of performance analysis and synthesis, a unified event-triggered Takagi-Sugeno (T-S) fuzzy switched model is established, which accounts for a suitable attack-resilient event-triggered communication scheme and unknown DoS jamming signals.Third, using piecewise Lyapunov-Krasovskii functional (PLKF) analysis technique, a newcriterion is derived to ensure exponential stability of the resulting switched tracking error system while achieving a weighted $H_{\infty}$ performance level. Additionally, the relationship among the parameters of a DoS attack signal, thetriggering parameters, the fuzzy controller gains, the sampling period, andthe decay rate can be quantitatively characterized. Moreover, the triggering matrix parameter and fuzzy controller gains are obtained by finding a feasible solution to a set of linear matrix inequalities (LMIs).Finally, numerical verification is performed to demonstrate the effectiveness of the proposed control design method.