Synchronization of Neural Networks Involving Distributed-Delay Coupling: A Distributed-Delay Differential Inequalities Approach.
Xiaoyu Zhang,Chuandong Li,Hongfei Li,Jing Xu
DOI: https://doi.org/10.1109/tnnls.2022.3224393
IF: 14.255
2024-01-01
IEEE Transactions on Neural Networks and Learning Systems
Abstract:In this article, we address the synchronization issue for coupled neural networks (CNNs) with mixed couplings by way of the delayed impulsive control, where the delay is distributed. Particularly, mixed couplings comprise the current-state coupling and the distributed-delay coupling, where influences on network connections caused by the past information of CNNs over a certain period are considered. First, we propose a novel array of delayed impulsive differential inequalities involving distributed-delay-dependent impulses, where distributed delays can be relatively larger. Second, we apply such delayed inequalities to analyze the problem of synchronization for CNNs with two different topologies. Sufficient criteria and distributed-delay-dependent impulsive controller are derived thereby. Furthermore, using techniques of matrix decomposition, several low-dimensional criteria are set out, which are appropriate for applications of large scale CNNs. Finally, a numerical example of CNNs with both the current-state coupling and the distributed-delay coupling involving three cases, are exhibited to exemplify the validity and the efficiency of the obtained theoretical results.