Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (36): 109-111.

• 网络、通信、安全 • Previous Articles     Next Articles

Synchronization judgment of interacting neural networks and its application to neural cryptography

TIAN Yong,XIANG Tao   

  1. Department of Computer Science,Chongqing University,Chongqing 400030,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-12-21 Published:2011-12-21


田 勇,向 涛   

  1. 重庆大学 计算机学院,重庆 400030

Abstract: Several scenarios of interacting neural networks especially tree parity machine are widely used in many fields such as cryptography.The judgment of time for neural synchronization is necessary.It provids an algorithm based on the hidden units of the neural networks and Hash functions which can do this kind of judgment safely,efficiently and fast.Amount of simulations are practiced in order to prove it.

Key words: neural networks, tree parity machine, synchronization, set of neurons

摘要: 互学习的神经网络特别是树状奇偶模型的神经网络因能通过一定量的信息交换达到同步而被广泛地应用在密码学等领域。提出树状奇偶机同步模型的同步判定的必要性和解决同步判断的算法,即基于树状奇偶机隐藏单元输出值的HASH值的比较,该算法将原同步算法的安全性改变到HASH函数的安全性和原同步算法的安全性之上,而在时间开销上也不会增加或增加得很少。仿真实验也证明了该算法判断同步所需要的时间复杂度较低。

关键词: 神经网络, 树状奇偶模型, 同步, 神经元集合