Event-triggered H_∞/passive synchronization for Markov jumping reaction-diffusion neural networks under deception...

Area of Science:

• Control Theory
• Networked Systems
• Applied Mathematics

Background:

• Master-slave synchronization is crucial for complex systems.
• Markov jumping neural networks (MJNNs) exhibit state-dependent dynamics.
• Reaction-diffusion terms introduce spatial complexity.
• Deception attacks pose significant security threats to control systems.

Purpose of the Study:

• To investigate H∞/passive master-slave synchronization for MJNNs with reaction-diffusion terms.
• To develop an event-triggered control scheme to reduce communication load.
• To address the challenge of deception attacks that alter control signals.

Main Methods:

• An event-triggered transmission scheme is designed to optimize communication bandwidth.
• Lyapunov stability theory is extended to derive sufficient conditions for synchronization.
• Matrix convex optimization is employed to determine controller gains.
• The proposed method accounts for randomly occurring deception attacks.

Main Results:

• Sufficient conditions for achieving the prescribed H∞/passive performance level are established.
• The event-triggered control scheme effectively manages communication bandwidth.
• The controller design handles deception attacks by considering sign modifications.
• The proposed approach is validated through a numerical example.

Conclusions:

• The developed event-triggered control scheme ensures robust H∞/passive synchronization for MJNNs under deception attacks.
• The methodology balances control performance with communication efficiency.
• The findings contribute to the secure and efficient control of complex networked systems.