STABILITY CRITERION OF A CLASS OF NEUTRAL NEURAL NETWORK WITH UNCERTAINTIES AND TIME-VARYING DELAY

Under reasonable postulated conditions of time-varying delay, the stability of a class of neutral neural network is probed. Based on the proper construction of the Lyapunov functional, a new method of predicating the uniformly asymptotic stability of this kind of neutral network is presented. The criterion is based on the positive definite solutions of a Riccati equation and a discrete Lyapunov equation, and is converted to LMI problem. The validity of the criterion is also confirmed by the simulation.