Nonfragile Synchronization of BAM Inertial Neural Networks Subject to Persistent Dwell-Time Switching Regularity

• Published in: IEEE transactions on cybernetics pp. 1 - 12
• Main Authors: Shen, Hao, Huang, Zhengguo, Wu, Zhengguang, Cao, Jinde, Park, Ju H.
• Format: Journal Article
• Language: English
• Published: IEEE 01.07.2022
• Subjects: Artificial neural networks; BAM inertial neural networks (BAMINNs); Biological neural networks; Control systems; Neurons; nonfragile H∞ synchronization; persistent dwell-time (PDT) switching regularity; Switched systems; Switches; Synchronization; time-varying delays (TVDs)
• ISSN: 2168-2267
• DOI: 10.1109/TCYB.2021.3119199

This article concentrates on the synchronization of discrete-time persistent dwell-time (PDT) switched bidirectional associative memory inertial neural networks with time-varying delays. Through the use of the switched system theory related to the PDT, the convex optimization technique together with some straightforward decoupling methods, an appropriate mode-dependent controller with nonfragility is developed to acclimatize itself to some practical circumstances. Simultaneously, sufficient conditions of ensuring the {H}_{∞} performance and exponential stability for the resulting switched synchronization error system are derived. Finally, a numerical example is utilized to show the validity of the model constructed and the influence of the PDT on the {H}_{∞} performance. In addition, an image encryption example is employed to show the potential application prospect of the investigated system.