Sampled-Data-Based H_∞ Synchronization of Switched Coupled Neural Networks.

Area of Science:

• Control Theory
• Neural Networks
• Systems Engineering

Background:

• Investigates sampled-data-based H-infinity synchronization for switched coupled neural networks.
• Addresses challenges of asynchronous switching and exogenous perturbations.
• Extends existing research beyond nonswitched and continuous-time systems.

Purpose of the Study:

• To develop sufficient conditions for sampled-data controllers in switched coupled neural networks.
• To ensure exponential stability of the synchronization error system.
• To limit the impact of external disturbances on synchronization performance.

Main Methods:

• Employs a time-dependent switching mechanism framework.
• Derives conditions for sampled-data controllers under variable sampling and asynchronous switching.
• Utilizes H-infinity control theory for disturbance attenuation.

Main Results:

• Guarantees exponential stability for the synchronization error system.
• Constrains the influence of exogenous perturbations to a specified level.
• Demonstrates applicability through simulations on switched coupled cellular and Hopfield neural networks.

Conclusions:

• The proposed sampled-data H-infinity control strategy is effective for switched coupled neural networks.
• The method successfully handles asynchronous switching and external disturbances.
• Validates the approach with practical neural network models.