H∞ state estimation for memristor-based neural networks with unbounded time-varying delays

This work is devoted to the problem of H
∞
state estimation for memristor-based neural networks with unbounded time-varying delays. The solution-based estimation method is proposed in this paper to obtain the criteria to ensure that the augmented system consisting of the dynamic equations of the memristor-based neural networks and the observer error system is globally exponentially stable with H
∞
performance level. This approach requires neither model transformation nor construction of Lyapunov–Krasovskii functionals. Finally, one numerical example and its numerical simulations are used to illustrate the applicability of the theoretical results.