Global robust stability analysis of uncertain neural networks with time varying delays

This paper deals with the global robust stability analysis of dynamical neural networks with time varying delays. By combining Lyapunov stability theorems and Homeomorphic mapping theorem, we obtain some original sufficient conditions for the existence, uniqueness and global asymptotic stability of the equilibrium point with respect to Lipschitz activation functions and under parameter uncertainties of the neural system. We also prove that the obtained robust stability conditions generalize some of the previously published corresponding literature results. The conditions we present can be easily verified as the conditions that are expressed in terms of the network parameters. Some comparative numerical examples are presented to demonstrate the advantages of our conditions over the previously published robust stability results.