Global exponential stability and periodic solutions of high-order bidirectional associative memory (BAM) neural networks with time delays and impulses

Abstract

High-order bidirectional associative memory (BAM) neural networks with mixed features such as time delays and impulses play an increasingly important role in the design and implementation of neural network systems. It exhibits advantage of stronger approximation property, faster convergence rate, greater storage capacity and higher fault tolerance than low-order neural networks. In this paper, issues of both stability and periodicity for a class of high-order BAM neural networks with time delays and impulses are investigated. With M-matrix theory, linear matrix inequality technique, fixed point theorem and Lyapunov approach, we derive new sufficient conditions for achieving global exponential stability of equilibrium point and the existence and global exponential stability of periodic solutions for the addressed high-order BAM neural networks. Numerical examples are provided to demonstrate the effectiveness of the proposed approach.