两类输入输出周期函数的离散神经网络的混沌  

Chaos in the Discrete-time Neural Networks with Two Types of Input-Output Period Functions

作　　者：阮炯[1] 

机构地区：[1]复旦大学数学科学学院,上海200433

出　　处：《徐州师范大学学报（自然科学版）》2008年第2期17-22,共6页Journal of Xuzhou Normal University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(39770707)

摘　　要：从理论上严格证明了以下结论:带有单峰与正反双向单峰周期输入输出函数的高阶瞬时混沌离散神经网络,在权阵W为非异常数矩阵,参数τ适当小,其它一些参数适当选取的条件下,该系统是Devaney意义下的混沌系统.所用的方法是反向可积极限方法与符号动力学方法的结合.By combination of anti-integrable limit method and note dynamics method,it is proved that for higher-order transiently chaotic discrete-time neural networks with positive direction single mont or positive-negative directions single mont of input-output period functions,suppose wight matrix is nonsingular constant matrix,parameter τ small,for other parameters some conditions are satisfied,then this system is chaotic in the sense of Devaney.

关 键 词：离散神经网络 混沌 输入输出周期函数 

分 类 号：O193[理学—数学]