摘要
掩码是当前抵御侧信道攻击的可证明安全类防护手段,但实际应用中,高阶掩码由于受到资源开销的限制,通常会和其他轻量化防护结合使用,如掩码与乱序组合的方案.现有掩码与乱序组合方案的安全性评估均基于特定的区分器或特定的攻击技术.本文从泄露量的角度不依赖于特定的区分器及攻击方法评估这种组合方式,相较于已有基于具体攻击技术的评估更基础且通用,评估结果不会因新型区分器的提出或攻击技术的革新而变化.在噪声模型下对实际泄露量采用噪声指标来进行量化,通过模型规约得到能够同时衡量掩码方案以及防护组合方案的安全水平下界.通过仿真与实测实验来分析防护组合方案中影响安全水平的可配置因素.实验结果证实,组合防护的噪声指标ARE与乱序阶数t呈反比,1阶掩码与乱序的组合防护能达到2阶或3阶掩码的攻击下界m的量级,即达到2阶或3阶安全水平.
Masking is the most widely employed provable secure countermeasure against side-channel analysis.However,when applying masking to protect the cryptographic algorithm,both resources and timing are at least as twice as those in the plain implementation.Load of area and speed limit the higher order masking in the real applications.Lightweight countermeasures are combined with masking to replace the high order masking for applications.The combination of masking and shuffling is considered as reasonable among researchers,and its resistance is based on different SCA distinguishers.This paper analyzes the combined scheme from the leakage perspective,which is independent of any specific analysis method.By modeling the physical leakage under the noisy model and reducing to random probing model,it is able to quantify the security level of any order masking schemes and the combined schemes.Some simulation experiments are conducted,and the results meet the theoretic conclusion.Experimental results show that the AREs are decreased in proportion to the order of shuffling t.Furthermore,the 1st-order masking combined with shuffling can reach the security of higher order masking as it has approximately the same magnitude of m,the minimal number of traces to successfully mount an attack.
作者
肖冲
唐明
严飞
XIAO Chong;TANG Ming;YAN Fei(Key Laboratory of Aerospace Information Security and Trusted Computing of Ministry of Education,School of Cyber Science and Engineering,Wuhan University,Wuhan 430072,China)
出处
《密码学报》
CSCD
2023年第1期155-167,共13页
Journal of Cryptologic Research
基金
国家自然科学基金(61972295)
武汉市科技项目应用基础前沿专项(2019010701011407)。
关键词
侧信道
掩码
乱序
轻量级防护
随机探针模型
side-channel analysis(SCA)
masking
shuffling
lightweight countermeasure
random probing model