H∞ state estimation for neural networks with mixed time delays

This paper studies the problem of H∞ state estimation for neural networks with mixed time-varying delays. Firstly, based on a newly augmented Lyapunov-Krasovskii functional (LKF), novel delay-dependent conditions are obtained such that the error system is globally asymptotically stable with H∞ performance index γ. Secondly, less conservative stable results are established by employing some effective mathematical techniques and Wirtinger integral inequality. Besides, new activation function conditions are proposed by introducing an adjustable parameter σ. The wishful estimator gain matrix can be formed in terms of linear matrix inequalities (LMIs). Finally, one numerical example with simulations is given to demonstrate the effectiveness and the advantage of the theoretical results.