基于改进型树型奇偶机的密钥交换研究  

Research on Key Exchange Based on Improved Tree Parity Machine

作　　者：李西明[1] 王璇 王浩[1] 陈志浩[1] 张民[2] 郭玉彬[1] LI Ximing;WANG Xuan;WANG Hao;CHEN Zhihao;ZHANG Min;GUO Yubin(School of Mathematics and Informatics,South China Agricultural University,Guangzhou 510642;School of Computer Science and Technology,Liaocheng University,Liaocheng 252000)

机构地区：[1]华南农业大学数学与信息学院,广州510642 [2]聊城大学计算机学院,聊城252000

出　　处：《计算机与数字工程》2024年第6期1721-1726,共6页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:61872152,61872409);2018年广东省农业厅省级乡村振兴战略专项项目(编号:粤农计(2018)54号);广东省基础与应用基础重大项目(编号:2019B030302008,2020A1515010751);广州市科技计划项目(编号:201902010081)资助。

摘　　要：提出了一种基于改进型树型奇偶机的密钥交换方法。利用伪随机数产生器为通信双方产生共同的输入向量,降低网络同步所需的数据量;使用基于滑动窗口的学习规则来动态改变学习率,降低同步次数;并且进一步提出基于背包算法和深度优先搜索遍历算法的高效输入序列生成方法,增大正向学习概率加快同步速度。仿真实验结果表明,改进模型的通信数据量下降了60%以上,同步时间显著下降。This paper proposes a key exchange method based on improved tree parity machine.The pseudo-random number generator is used to generate common input vectors for both sides of communication,so as to reduce the amount of data needed in network synchronization.Sliding window based learning rules are used to dynamically change the learning rate,so as to reduce the number of synchronization.Furthermore,an efficient input sequence generation method based on knapsack algorithm and depth first search traversal algorithm is proposed to increase the positive learning probability and accelerate the synchronization speed.The sim-ulation results show that the communication data volume of the improved model is reduced by more than 60%,and the synchroniza-tion time is significantly reduced.

关 键 词：密钥交换 网络同步 互学习 树型奇偶机(TPM) 

分 类 号：TU205[建筑科学—建筑设计及理论]