State estimation for discrete-time fractional-order neural networks with time-varying delays and uncertainties

In this paper, the state estimation issue for a class of discrete-time fractional-order neural networks (DFNNs) with time-varying delays and uncertainties is investigated. Based on the theoretical results of MittagLeffler function and Caputo fractional difference, a novel discrete-time inequality is established, which has generality compared with the previous results. By designing suitable estimator, exploiting Lyapunov method and combining inequality we establish, sufficient conditions ensuring the global asymptotical stability of estimation error system are obtained through linear matrix inequalities (LMIs). Moreover, we further discuss the case of DFNNs without uncertainties. In the end, numerical examples are utilized to validate availability of our theoretical results.