Dynamic Event-Triggered Passive Synchronization for Mode-Dependent Persistent Dwell-Time Switched Neural Networks Subject to DoS Attacks

• Published in: Neural processing letters Vol. 55; no. 9; pp. 12421 - 12436
• Main Authors: Wang, Xiaomin, Xia, Jianwei, Su, Lei, Shen, Hao
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.12.2023; Springer Nature B.V
• Subjects: Artificial Intelligence; Communication; Complex Systems; Computational Intelligence; Computer Science; Controllers; Criteria; Data transmission; Denial of service attacks; Dwell time; Neural networks; Stability; Subsystems; Synchronism; Neural networks; Mode-dependent persistent dwell-time; Synchronization control; Dynamic event-triggered mechanism; Activation function dividing method
• ISSN: 1370-4621; 1573-773X
• DOI: 10.1007/s11063-023-11426-7

The synchronization control problem of discrete-time switched neural networks is considered in this paper. Wherein, the switchings among subsystems are described by the mode-dependent persistent dwell-time switching rule. Due to restricted network bandwidth resources, a dynamic event-triggered mechanism is introduced to alleviate the frequency of data transmission. In addition, the activation function dividing method is utilized to make the result less conservative. Then, using Lyapunov stability theory, several sufficient criteria are obtained to ensure that the synchronization error system can achieve mean-square exponential stability and meet the specified passive performance under DoS attacks. Based on these criteria, the concrete form of controller gain is solved. Finally, numerical examples demonstrate the rationality and superiority of the proposed method.