具比例时滞杂交双向联想记忆神经网络的全局指数稳定性  

Global Exponential Stability in Hybrid Bidirectional Associative Memory Neural Networks with Proportional Delays

作　　者：周立群[1] 

机构地区：[1]天津师范大学数学科学学院,天津300387

出　　处：《电子学报》2014年第1期96-101,共6页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61374009);天津市高等学校科技发展基金(No.20100813);天津师范大学博士基金(No.52LX34)

摘　　要：对一类具比例时滞杂交双向联想记忆神经网络进行研究,利用Brouwer不动点定理证明该网络的平衡点的存在唯一性.利用变换将具比例时滞杂交双向联想记忆神经网络变换成等价的具不等常时滞与变系数杂交双向联想记忆神经网络.利用不等式技巧建立一拟Halanay型不等式系统,进而得到了确保该系统全局指数稳定的时滞独立的充分条件.并给出两个算例验证所得结论的正确性.In this paper,a class of hybrid bidirectional associative memory neural networks with proportional delays is stud- ied. In virtue of Bmuwer fixed point theorem, the existence and uniqueness of equilibrium of the system are proved. The transforma- tion Iransforms the neural networks with proportional delays into the neural networks with unequal constant delays and variable coef- ficients.By applying inequality techniques, a system of the quasi Halanay-type inequalities is established, and a delay-independent sufficient condition is derived for ensuring the global exponential stability of equilibrium of the system. And two examples are given to illuslrate the effectiveness of obtained results.

关 键 词：联想记忆 神经网络 比例时滞 BROUWER不动点定理 全局指数稳定性 

分 类 号：O175.13[理学—数学] TP183[理学—基础数学]