论文部分内容阅读
SMS4算法一种是用于WAPI的分组密码算法,也是国内官方公布的第一个商用密码算法,该算法公布后即引起国内外密码学界的分析热潮.SMS4算法的分组长度为128比特,密钥长度为128比特,加密算法与密钥扩展算法都采用32轮迭代结构.本文的分析方法是综合利用了2~(28)个17轮的SMS4的差分特征,采用基于最优区分器思想的多差分攻击方法对21轮的SMS4算法进行攻击和分析,针对每个实验密钥,构造出基于多个差分特征的统计量,根据统计量的大小判决实验密钥是否是正确密钥.给出了多差分分析方法的计算复杂度,分析了正确密钥、错误密钥对应统计量的概率分布规律,在此基础上给出了多差分分析方法的成功率和数据复杂度之间的关系.最终得出结论可以2~(104)的数据复杂度,2~(114)的计算复杂度,来恢复出该算法的128比特圈子密钥.用该结果与目前已知的对21轮SMS4算法的差分攻击结果进行对比我们可以看出,攻击的数据复杂度和计算复杂度都有所降低.基于该研究结果,我们可以得出以下结论,在成功率相同的条件下,基于的差分特征越多,需要的数据复杂度和计算复杂度越小.
The SMS4 algorithm is a block cipher algorithm used in WAPI and also the first official commercial cipher algorithm published in China, which has caused the analysis craze in the field of cryptology at home and abroad since its publication.SMS4 algorithm has a packet length of 128 bits and the key The length of 128 bits, encryption algorithm and key expansion algorithm using 32 rounds of iterative structure.This method is the comprehensive analysis of the use of 2 to (28) 17 rounds of SMS4 difference characteristics, based on the optimal divider The differential attack method attacks and analyzes 21-round SMS4 algorithm, constructs statistics based on multiple differential features for each experimental key, and determines whether the experimental key is the correct key according to the size of the statistics. The computational complexity of multi-difference analysis method is analyzed, and the probability distribution law of the correct key and error key corresponding statistics is analyzed. Based on this, the relationship between the success rate of multi-difference analysis and data complexity is given. The conclusion is that the 128-bit circle key of this algorithm can be restored with the complexity of 2 ~ (104) and the computational complexity of 2 ~ (114) .This result is consistent with the known 21 pairs of SMS4 algorithms Differential attack We can see from the comparison of the results that both the data complexity and computational complexity of the attacks are reduced.According to the results of this study, we can draw the following conclusion: Based on the same success rate, the more differential features are needed, The less data complexity and computational complexity.