Adaptive synchronization of a class of chaotic neural networks with time-varying delays and uncertain parameters

This paper is concerned with the asymptotic synchronization of a class of time-varying delayed chaotic neural networks with parameter uncertainties. Using the drive-response concept, in terms of a linear matrix inequality (LMI) and the Lyapunov stability theory, two sufficient conditions for global asymptotic synchronization of uncertain chaotic delayed neural networks are derived under the differentiable and non-differentiable conditions of time-varying delays respectively, which also present a procedure to construct synchronization controllers. Under the non-differentiable condition of time-varying delays, the sufficient condition generalizes and further improves those in the earlier publications. The examples are given to demonstrate the effectiveness of the present method.