Hybrid-triggered and fault-tolerant observer-based control for neural networks under malicious attacks

Abstract

Based upon hybrid-triggered mechanism, this article inspects fault-tolerant issue based (Q,S,R) dissipative control for neural networks system along with malcious attacks and external disturbances. To be precious, conventional Luenberger observer is employed to estimate the system state and malicious attack signal, whereas the presence of smoothed signal of malicious attack in the system state will not allow the augmented state to be precisely estimate. Moreover, the hybrid-triggered mechanism which incorporates both time- and event-triggered scheme is initiated to mitigate the redundancy of network transmission and secures the network resources. Preciously, the stochastic switching within the mechanism satisfies the Bernoulli distribution. The main intention of this study is to develop a hybrid-triggered scheme and a fault-tolerant controller for ensuring the mean-square asymptotic stability for the desired neural networks with (Q,S,R) dissipative performance index. Assisted by Lyapunov stability theory and some inequality techniques, an adequate criterion has been attained in the configuration of linear matrix inequalities (LMIs) to guarantee the asymptotic stability of the neural networks system. Further, the anticipated gain matrices are attained with the strength of obtained LMIs. In the final analysis, the applicability and efficacity of the proposed control model are reflected through two numerical examples.