Finite-Time Boundedness of Impulsive Delayed Reaction–Diffusion Stochastic Neural Networks

TY - JOUR

T1 - Finite-Time Boundedness of Impulsive Delayed Reaction–Diffusion Stochastic Neural Networks

AU - Yao, Qi

AU - Wei, Tengda

AU - Lin, Ping

AU - Wang, Linshan

N1 - Copyright: © 2024 IEEE.

PY - 2024/2/22

Y1 - 2024/2/22

N2 - Considering the impulsive delayed reaction–diffusion stochastic neural networks (IDRDSNNs) with hybrid impulses, the finite-time boundedness (FTB) and finite-time contractive boundedness (FTCB) are investigated in this article. First, a novel delay integral inequality is presented. By integrating this inequality with the comparison principle, some sufficient conditions that ensure the FTB and FTCB of IDRDSNNs are obtained. This study demonstrates that the FTB of neural networks with hybrid impulses can be maintained, even in the presence of impulsive perturbations. And for a system that is not FTB due to impulsive perturbations, achieving FTB is possible through the implementation of appropriate impulsive control and optimization of the average impulsive intervals. In addition, to validate the practicality of our results, three illustrative examples are provided. In the end, these theoretical findings are successfully applied to image encryption.

AB - Considering the impulsive delayed reaction–diffusion stochastic neural networks (IDRDSNNs) with hybrid impulses, the finite-time boundedness (FTB) and finite-time contractive boundedness (FTCB) are investigated in this article. First, a novel delay integral inequality is presented. By integrating this inequality with the comparison principle, some sufficient conditions that ensure the FTB and FTCB of IDRDSNNs are obtained. This study demonstrates that the FTB of neural networks with hybrid impulses can be maintained, even in the presence of impulsive perturbations. And for a system that is not FTB due to impulsive perturbations, achieving FTB is possible through the implementation of appropriate impulsive control and optimization of the average impulsive intervals. In addition, to validate the practicality of our results, three illustrative examples are provided. In the end, these theoretical findings are successfully applied to image encryption.

KW - Delay effects

KW - Delays

KW - finite-time boundedness (FTB)

KW - finite-time contractive boundedness (FTCB)

KW - impulses

KW - Neural networks

KW - Optimization

KW - Perturbation methods

KW - reaction–diffusion stochastic neural networks

KW - Stochastic processes

KW - Sufficient conditions

U2 - 10.1109/TNNLS.2024.3360711

DO - 10.1109/TNNLS.2024.3360711

M3 - Article

C2 - 38386575

SN - 2162-237X

VL - 36

SP - 4794

EP - 4804

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

IS - 3

M1 - 10443721

ER -