A Fast and Accurate Guessing Entropy Estimation Algorithm for Full-key Recovery
A Fast and Accurate Guessing Entropy Estimation Algorithm for Full-key Recovery
复制标题
一种快速准确的全密钥恢复猜测熵估计算法
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Yunsi Fei
中科院分区：
文献类型：
--
作者：
Ziyue Zhang;A. Ding;Yunsi Fei
猜测熵（GE）是一种广泛采用的度量标准，用于衡量成功的侧信道分析（SCA）所需的平均计算成本。然而，在当前的估计方法中，评估者必须在许多独立的侧信道泄漏测量集上平均正确的密钥秩，全密钥GE估计由于其过高的计算要求而不切实际。针对现有的基于后验概率的估计方法虽然具有可扩展性，但并不准确的问题，提出了一种新的基于排序得分向量理论分布的猜测熵估计算法（GEEA）。GEEA通过发现GE与两两成功率的关系，并利用它，使用多个单变量高斯概率之和代替多变量高斯概率，显著提高了计算效率，比现有的GE估计更准确、更有效。据我们所知，这是唯一实用的全键GE评价给定的实验数据集，评估人员可以访问。此外，它可以准确地预测GE比实验数据集更大的规模，提供全面的安全评估。
Guessing entropy (GE) is a widely adopted metric that measures the average computational cost needed for a successful side-channel analysis (SCA). However, with current estimation methods where the evaluator has to average the correct key rank over many independent side-channel leakage measurement sets, full-key GE estimation is impractical due to its prohibitive computing requirement. A recent estimation method based on posterior probabilities, although scalable, is not accurate.We propose a new guessing entropy estimation algorithm (GEEA) based on theoretical distributions of the ranking score vectors. By discovering the relationship of GE with pairwise success rates and utilizing it, GEEA uses a sum of many univariate Gaussian probabilities instead of multi-variate Gaussian probabilities, significantly improving the computation efficiency.We show that GEEA is more accurate and efficient than all current GE estimations. To the best of our knowledge, it is the only practical full-key GE evaluation on given experimental data sets which the evaluator has access to. Moreover, it can accurately predict the GE for larger sizes than the experimental data sets, providing comprehensive security evaluation.