Exponential stability of delayed neural networks with delayed sampled-data inputs: An extended bilateral looped functional approach

Abstract

In this paper, the exponential stability of delayed neural networks (DNNs) with delayed sampled-data inputs is investigated via extended bilateral looped functional approach. Firstly, a new extended bilateral looped functional is constructed, which is differentiable at sampling intervals and can relax the constraints on positive definiteness when compared to traditional functionals. Then, less conservative criteria for exponential stability of DNNs with delayed sampled-data inputs expressed through linear matrix inequalities (LMIs) are obtained. Furthermore, the results are extended to T–S fuzzy DNNs with delayed sampled-data inputs, where corresponding stability conditions are likewise derived. Finally, two simulation examples are given to illustrate the validity of the main results.