Advanced stability criteria for static neural networks with interval time-varying delays via the improved Jensen inequality

Abstract

This paper investigates the asymptotic stability of static neural network (SNN) with interval time-varying delay. A proper Lyapunov-Krasovskii functional (LKF) is presented to coordinate with the features of the improved Jensen inequality. Based on the new proposed LKF together with the improved Jensen inequality and a parameter-dependent matrix inequality, new sufficient conditions ensuring the delay-dependent stability are derived in terms of linear matrix inequality (LMI). By considering a marginal augmented vector with two complementary integral terms to yield more crossing information, a further enhanced stability criterion is obtained. The validity of the theorectical results is demonstrated by two numerical examples.