期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2006
卷号：6
期号：2B
页码：63-63~69
出版社：International Journal of Computer Science and Network Security
摘要：In this paper, the periodic solutions have been discussed for a general class of bi-directional associative memory (BAM) neural networks with time-varying delays and reaction-diffusion terms. A simple and new sufficient condition has been obtained to ensure the existence of periodic solutions of bi-directional associative memory neural networks with time-varying delays and reaction-diffusion terms, and the activation functions are not required to be differentiable, bounded, monotone nondecreasing and the weight-connected matrices to be symmetric in the obtained result. In addition, an example has been given to show the effectiveness of the obtained result.
