具有时滞的Hopfield神经网络的周期解

A Periodic Solution of Hopfield Neural Network with Delays

  • 摘要: 假设具有时滞的Hopfield神经网络的每个输出响应函数是满足Lipschitz条件的有界函数,当该网络的输入信号始终以正常数ω为周期,并且在网络参数满足一定的条件时,通过构造适当的Lyapunov泛函的方法,得到了该类网络必存在唯一的ω-周期解,并且其余各解都按指数收敛于该周期解的一些判据,通过实例对所得到的判据进行了直观性解释。

     

    Abstract: Some sufficient criterias about the existence and uniqueness of a ω-periodic solution are obtained under the assumption of that each output response is bounded and satisfies Lipschitz condition for Hopfield neural network, when all input signals are continuously ω-periodic functions and the network parameters satisfy suitable conditions, by means of the method of a suitable Lyapunov functional. And it is proved that all other solutions converge exponentially to the above ω-periodic solution. An intuitive explain of the above new criteras is given in the final example.

     

/

返回文章
返回