AES密码芯片DPA攻击的关联检测算法设计

Design of Association Detection Algorithm for DPA Attacks on AES Cryptographic Chip

当芯片在执行不同的指令各种运算时,对应的功率消耗也相应变化。DPA攻击使用数学统计技术,来检测和分析这些变化,从而得到芯片中的特定关键信息,对这种攻击准确检测至关重要。提出了基于关联规则的AES密码芯片DPA攻击检测算法。利用关联规则算法,获取加密数据的关联度主特征量,根据AES密码芯片遭受的单分量信号模型,获取DPA攻击信号的伸缩变化,通过对攻击信号的分解,结合尺度仿射变换,实现了AES密码芯片DPA攻击的检测。实验结果表明:所提方法能够准确检测到AES密码芯片遭受的DPA攻击,降低了误检率,提高了AES密码芯片的安全性。

When the chip performs various operations with different instructions,the corresponding power consumption changes accordingly.DPA attacks use mathematical statistics technology to detect and analyze these changes,so as to obtain specific key information in the chip,which is crucial for accurate detection of such attacks.A DPA attack detection algorithm for AES cryptographic chip based on association rules is proposed.The association rule algorithm is used to obtain the main characteristic quantity of the association degree of the encrypted data.According to the single component signal model suffered by the AES cryptographic chip,the scaling change of the DPA attack signal is obtained.Through the decomposition of the attack signal,combined with the scale affine transformation,the detection of the AES cryptographic chip DPA attack is realized.The experimental results show that the proposed method can accurately detect the DPA attacks on AES cryptographic chips,reduce the false detection rate,and improve the security of AES cryptographic chips.