Abstract: The non-fragile state estimation problem is investigated for a class of continuous neural networks with time-delays and nonlinear perturbations. As the estimator parameters cannot be accurately obtained in practice, a non-fragile state estimator with additive bounded gain uncertainty is used to describe this phenomenon. The main purpose of the addressed problem is to design a non-fragile state estimator for the delayed neural networks such that the dynamics of the estimation error converges to equilibrium asymptotically. Utilizing a combination of the Lyapunov functionals and the matrix analysis techniques, sucient conditions are established for the asymptotic stability, and the gain characteristics of the state estimator are obtained. Finally, a simulation example is used to illustrate the effectiveness of the proposed method.