Some Sufficient Conditions for Global Exponential Stability of Delayed Hopfield Neural Networks

Ht Lu,Fl Chung,Zy He

DOI: https://doi.org/10.1016/j.neunet.2004.01.004

IF: 7.8

2004-01-01

Neural Networks

Abstract:In this paper, we have derived some sufficient conditions for existence and uniqueness of equilibrium and global exponential stability in delayed Hoplield neural networks by using a different approach from the usually used one where the existence, uniqueness of equilibrium and stability are proved in two separate steps, rather we first prove global exponential convergence to 0 of the difference between any two solutions of the original neural networks, the existence and uniqueness of equilibrium is the direct results of this procedure. We obtain the conditions by suitable construction of Lyapunov functionals and estimation of derivates of the Lyapunov functionals by the well-known Young's inequality and Holder's inequality. The proposed conditions are related to p-norms of vector or matrix, p is an element of [ 1, infinity], and thus unify and generalize some results in the literature. (C) 2004 Published by Elsevier Ltd.