Dynamic event-triggered synchronization control for neutral-type SMJ neural networks with additive delays under synchronized attacks

Abstract

This paper studies the double event-triggered synchronization (ETS) of neutral-type semi-Markovian jump (SMJ) neural networks under synchronous attacks. Firstly, synchronous attacks are modeled by an independent semi-Markovian jump process. Secondly, the double dynamic event-triggered mechanisms (DDETMs) introduced offer the advantage of conserving communication resources and alleviating the computational burden. Thirdly, considering that both the network of sensor to controller and controller to actuator may cause time-varying delays (TVDs), and additive TVDs are designed to enhance the dynamics affected by these delays in the network. Subsequently, a semi-Markov dynamic event-triggered controller is designed to ensure the synchronization of neutral-type SMJ neural networks. Then, the conservatism of the synchronization criterion is reduced by using methods such as asymmetric Lyapunov–Krasovskii functions (LKFs) and novel reciprocally convex combination inequality (RCCI). Finally, two sets of values are given to prove the effectiveness of the synchronization criterion of the proposed method.