State Estimation for Discrete Time-Delayed Impulsive Neural Networks Under Communication Constraints: A...

Area of Science:

• Control Theory
• Artificial Intelligence
• Neural Networks

Background:

• State estimation for delayed impulsive neural networks presents significant challenges.
• Existing methods often lack generality in activation functions and efficient data handling.

Purpose of the Study:

• To develop a delay-range-dependent approach for state estimation in delayed impulsive neural networks.
• To design a state observer ensuring asymptotic stability of estimation error dynamics.
• To utilize a generalized nonlinear activation function and a round-robin protocol for enhanced performance.

Main Methods:

• A delay-range-dependent approach is employed for state estimation.
• Lyapunov stability theory is used to construct a stable state observer.
• Linear matrix inequalities (LMIs) are utilized for observer synthesis.
• A round-robin protocol mitigates network congestion.

Main Results:

• A novel nonlinear activation function, more general than sigmoid and Lipschitz functions, is adopted.
• The proposed observer guarantees asymptotic stability of the estimation error dynamics.
• Delay-range-dependent criteria ensure observer existence.
• Simulations validate the effectiveness of the proposed approach.

Conclusions:

• The developed delay-range-dependent method effectively addresses state estimation for delayed impulsive neural networks.
• The approach ensures stability and improves communication efficiency.
• The generalized activation function and round-robin protocol contribute to robust performance.