期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
An Improved Differential Fault Analysis on Block Cipher KLEIN-64
1
作者 Min Long Man Kong +1 位作者 Sai Long Xiang Zhang 《Computers, Materials & Continua》 SCIE EI 2020年第11期1425-1436,共12页
KLEIN-64 is a lightweight block cipher designed for resource-constrained environment,and it has advantages in software performance and hardware implementation.Recent investigation shows that KLEIN-64 is vulnerable to ... KLEIN-64 is a lightweight block cipher designed for resource-constrained environment,and it has advantages in software performance and hardware implementation.Recent investigation shows that KLEIN-64 is vulnerable to differential fault attack(DFA).In this paper,an improved DFA is performed to KLEIN-64.It is found that the differential propagation path and the distribution of the S-box can be fully utilized to distinguish the correct and wrong keys when a half-byte fault is injected in the 10th round.By analyzing the difference matrix before the last round of S-box,the location of fault injection can be limited to a small range.Thus,this improved analysis can greatly improve the attack efficiency.For the best case,the scale of brute-force attack is only 256.While for the worst case,the scale of brute-force attack is far less than 232 with another half byte fault injection,and the probability for this case is 1/64.Furthermore,the measures for KLEIN-64 in resisting the improved DFA are proposed. 展开更多
关键词 Block cipher klein-64 differential fault analysis half-byte fault injection
下载PDF
基于ResNet的KLEIN算法改进模板攻击 被引量:1
2
作者 王永娟 王灿 +1 位作者 袁庆军 冯芯竹 《密码学报》 CSCD 2022年第6期1028-1038,共11页
针对传统模板攻击存在的多元高斯正态分布假设受限、预处理复杂度高且不适用于带掩码防护的应用场景等问题,研究基于深度学习的模板攻击的改进方法.利用深度学习模型ResNet,对轻量级分组密码算法KLEIN实施改进模板攻击,根据数据的标签... 针对传统模板攻击存在的多元高斯正态分布假设受限、预处理复杂度高且不适用于带掩码防护的应用场景等问题,研究基于深度学习的模板攻击的改进方法.利用深度学习模型ResNet,对轻量级分组密码算法KLEIN实施改进模板攻击,根据数据的标签对数据进行分类.在密钥恢复阶段利用密钥优势叠加的方法,平均需要15条相同密钥加密所产生的能量迹即可有效区分正确密钥.相较于传统的模板攻击,本文的攻击方法成功恢复密钥所需攻击能量迹减少了83.7%,降低了模板攻击的难度,有效提高了模板攻击的成功率和效率. 展开更多
关键词 模板攻击 klein-64算法 深度学习 ResNet
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部