Exponential stability criterion of high-order BAM neural networks with delays and impulse via fixed point approach

• Published in: Neurocomputing (Amsterdam) Vol. 292; pp. 63 - 71
• Main Authors: Pu, Zhilin, Rao, Ruofeng
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 31.05.2018
• Subjects: Contraction mapping theory; Exponential stability; High-order bidirectional associative memory (BAM) neural networks; Exponential stability; Contraction mapping theory; High-order bidirectional associative memory (BAM) neural networks
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2018.02.081

High-order BAM neural networks have always been investigated by means of Lyapunov function method. However, in this paper, Banach fixed point theory and technique is employed to derive a new stability criterion of high-order BAM neural networks with delays and impulse for the first time. It is worth mentioning that using Banach fixed point theorem and technique results in deleting a constraint condition on the allowable upper bounds of time delays of existing results. And it is the utilization of Banach fixed point theorem that makes us recognize that the stability of the high order BAM neural networks is independent of the limit of time delay if active functions are bounded. Moreover, a numerical example is presented to illustrate the effectiveness of the proposed methods.