计算机与现代化

• 网络与通信 • 上一篇    下一篇

一种改进的基于树形奇偶机的神经网络同步方案

  

  1. 重庆大学计算机学院,重庆400044
  • 收稿日期:2014-03-31 出版日期:2014-05-28 发布日期:2014-05-30
  • 作者简介:刘忆璐(1989-),女,四川自贡人,重庆大学计算机学院硕士研究生,研究方向:信息安全,神经密码; 廖晓峰(1964-),男,教授,博士,研究方向:混沌加密,非线性控制理论; 梁一峰(1987-),男,硕士研究生,研究方向:信息安全,人工神经网。
  • 基金资助:
    国家自然科学基金资助项目(61170249)

An Improved Scheme for Neural Synchronization Based on Tree Parity Machine

  1. College of Computer Science, Chongqing University, Chongqing 400044, China
  • Received:2014-03-31 Online:2014-05-28 Published:2014-05-30

摘要: 神经网络通过相互学习达到完全同步来构建密码协议,已经成为当今密码学的重要研究方向。针对神经密码同步过程中通信次数过多的问题,本文利用神经密码应用中的树形奇偶机,在综述神经密码协议研究的基础上,提出对神经网络单元初始权值的选取进行改进的解决方案。仿真实验结果表明,在保证安全性的情况下,本文方案大大加快了同步速度。

关键词: 树形奇偶机, 神经网络同步, 密钥交换, 神经密码

Abstract: Neural network can fully synchronize by learning from each other, because this effect neural synchronization can be used to construct a cryptographic key-exchange protocol, which has become an important research direction in current cryptography. To solve the problem of too many times in neural cryptography synchronization, an improved scheme was proposed by employing Tree Parity Machine (TPM). On the basis of neural cryptographic protocols, the range of initial weights of neural units was appropriately reduced. After the analysis, the simulation results show that in the case of ensuring security, synchronization efficiency is greatly improved by applying new improvement scheme.

Key words: tree parity machine, neural synchronization, key-exchange, neural cryptography

中图分类号: