J4

• 研究论文 • 上一篇    下一篇

延迟离散Hopfield神经网络的动态行为

马润年1,3;雷社平2;郑连清3;许进1   

  1. (1. 西安电子科技大学 电子工程研究所,陕西 西安 710071;
    2. 西安理工大学 水利水电学院,陕西 西安 710048;
    3. 空军工程大学 电讯工程学院,陕西 西安 710077)

  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2003-04-20 发布日期:2003-04-20

Dynamic behavior of discrete Hopfield neural networks with delay

Ma Run-nian1,3;LEI She-ping2;ZHENG Lian-qing3;XU Jin1

  

  1. (1. Research Inst. of Electronic Engineering, Xidian Univ., Xi’an 710071, China;
    2. Inst. of Water Resource and Hydro-electric Eng., Xi'an Univ. of Tech., Xi'an 710048, China;
    3. The Telecommunication Eng. Inst., Air Force Eng. Univ., Xi'an 710077, China)

  • Received:1900-01-01 Revised:1900-01-01 Online:2003-04-20 Published:2003-04-20

摘要: 主要研究带有延迟项的离散Hopfield神经网络动态演化行为. 给出了延迟网络并行收敛于周期小于等于2的极限环的一些条件. 同时,得到了延迟网络收敛于周期为2的极限环的条件. 而且对所获结果通过例子进行了验证. 这些结果为延迟离散Hopfield神经网络的进一步分析和应用提供了一定的理论基础.

关键词: 延迟离散Hopfield网络, 稳定点, 极限环

Abstract: The dynamic behavior of discrete Hopfield neural networks with delay is studied. The conditions that the networks with delay are convergent to a limit cycle of length 1 or 2 are given. Also, the conditions that the networks with delay are convergent to a limit cycle of length 2 are achieved. Furthermore, the obtained results are validated by giving examples. These results here provide a theoretical foundation for further analysis and new applications of the discrete Hopfield neural networks with delay.

Key words: discrete Hopfield neural network with delay, stable state, cycle

中图分类号: 

  • TP183