Hybrid event-triggered network-based synchronization of MSRDNNs with additive mode-dependent time-varying delays

• Published in: The Journal of supercomputing Vol. 81; no. 15; p. 1427
• Main Authors: Zhang, Weiyuan, Li, Junmin, Zhang, Rui, Xing, Keyi
• Format: Journal Article
• Language: English
• Published: New York Springer US 09.10.2025; Springer Nature B.V
• Subjects: Artificial intelligence; Communication; Compilers; Computer Science; High performance computing; Interpreters; Network control; Neural networks; Ordinary differential equations; Processor Architectures; Programming Languages; Random variables; Sampling; Signal processing; Synchronism; Time dependence; Additive time-varying delays; Switching approach; Distributed spatiotemporal event-triggering synchronization; MSRDNNs
• ISSN: 1573-0484; 0920-8542; 1573-0484
• DOI: 10.1007/s11227-025-07926-z

Synchronization of neural networks (NNs) finds applications in various fields such as optimized intelligent computation, image encryption and information science. In this paper, a switching-based distributed spatiotemporal event-triggered method is proposed to address the synchronization control issue for Markovian switching neural networks subject to reaction–diffusion terms and additive mode-dependent time delays. The synchronization scheme focused on networked control, where the network-induced delays and spatiotemporal sampling are occurred. In addition, the delays are admitted to be longer than time sampling intervals. Combining input delay and dynamic delay interval ideas, by means of Wirtinger’s inequality, free weighting matrices and mutually convex combination techniques, synchronization criteria LMIs based are established, which can guarantee the synchronization of state trajectories for the considered NNs via a switching-based event-triggered approach under spatiotemporal sampling. The given examples are shown to support efficiency and potential synchronization mechanism.