Abstract
In this paper, the H infinity state estimation problem is investigated for a class of uncertain discrete-time neural networks subject to infinitely distributed delays and fading channels. Randomly occurring uncertainties (ROUs) are introduced to reflect the random nature of the network condition fluctuations, and the channel fading phenomenon is considered to account for the possibly unreliable network medium on which the measurement signal is transmitted. A set of Bernoulli-distributed white sequences are employed to govern the ROUs and the L-th Rice fading model is utilized where channel coefficients are mutually independent random variables with certain probability density function on [0, 1]. We aim to design a state estimator such that the dynamics of the estimation error is asymptotically stable while satisfying the prescribed H infinity performance constraint. By adopting the Lyapunov-Krasovskii functional and the stochastic analysis theory, sufficient conditions are established to ensure the existence of the desired state estimators and the explicit expression of such estimators is acquired. A simulation example is provided to verify the usefulness of the proposed approach. (C) 2016 Elsevier Ltd. All rights reserved.