一种基于奇异值分解的功耗轨迹筛选方法

发布时间：2018-12-24 10:14

【摘要】：功耗分析攻击是侧信道分析中针对密码设备最有效的分析手段之一,它利用密码设备消耗的功耗来分析密码设备的敏感信息.差分功耗分析是最早提出的功耗分析方法,也是目前最基本的分析方法之一.但是在实际使用差分功耗分析过程中,由于功耗轨迹存在噪声等因素,往往使得花了较多的功耗轨迹,差分功耗分析的效果一般,难以恢复出正确密钥.针对这个问题,本文提出了一种基于奇异值分解的选择功耗轨迹方法,这种方法可以选择一些质量好的功耗轨迹用于差分功耗分析,提高差分功耗分析的攻击效率.本文的实验验证了该方法的有效性,在同等分析条件下,对于我们自己采集的功耗数据,使用该方法情况下仅需124条功耗轨迹就可以达到80%的成功率,而普通差分功耗分析需要490条;对于DPA Contest 2008/2009提供的数据,使用该方法仅需53条功耗轨迹可以达到80%的成功率,而普通差分功耗分析需要195条.两个不同的实验对象都说明了该方法的有效性.
[Abstract]:Power analysis attack is one of the most effective analysis methods for cryptographic devices in side-channel analysis. It uses the power consumption of cryptographic devices to analyze sensitive information of cryptographic devices. Differential power analysis (DPA) is the first proposed power analysis method and one of the most basic analysis methods. However, in the practical use of differential power analysis, due to noise and other factors in the power path, it often spends more power trajectories, and the effect of differential power analysis is general, so it is difficult to recover the correct key. In order to solve this problem, this paper presents a method of selecting power path based on singular value decomposition. This method can select some high-quality power trajectories for differential power analysis and improve the attack efficiency of differential power analysis. The experiments in this paper verify the effectiveness of the method. Under the same analysis conditions, the power consumption data collected by ourselves can achieve a success rate of 80% in only 124 power consumption trajectories. The ordinary differential power analysis requires 490; For the data provided by DPA Contest 2008 / 2009, using this method requires only 53 power trajectories to achieve a success rate of 80%, while the conventional differential power analysis requires 195. Two different experimental objects show the effectiveness of the method.
【作者单位】：中国科学院信息工程研究所;中国科学院大学网络空间安全学院;
【基金】：国家自然科学基金项目(No.61372062)
【分类号】：TP309

【相似文献】