On robust exponential stability of delayed neural networks

The problems of the global robust exponential stability of interval delayed neural networks (IDNN) are considered. Based on a new matrix inequality, an approach combining the Lyapunov-Krasovskii functional with the linear matrix inequality is taken to investigate the problems. Some new conditions are obtained to ensure the existence, uniqueness and global robust exponential stability of the equilibrium point of IDNN with globally Lipschitz continuous activation functions. In particular, the exponential convergence rate for IDNN is estimated in terms of the proposed stability criteria. The effects of the time delays on the exponential convergence rate are also analyzed in detail.