摘要
针对芯片生产链长、安全性差、可靠性低,导致硬件木马防不胜防的问题,提出一种改进的机器学习分类算法.首先采集不同电压下电路的延时信号,通过KNN分类算法找出延时差异,若延时与干净电路相同,则判定为干净电路,否则判定有木马;然后联合多项式回归算法对木马延时特征进行拟合,基于回归函数建立木马特征库,最终实现硬件木马的准确识别.实验结果表明,对2 000组延时单元的19个不同电压进行延时提取,同时考虑电压数目、K值与识别准确率,则电压数目与木马的识别准确率成正比,而参数K与识别准确率成反比;综合考虑的电压数目为19时,其预测准确率达到最高的95.2%;所提算法能明显地提升硬件木马的识别准确率和自动化程度.
Hardware Trojan detection is too difficult due to the reason that chip production chain is too long. An improved machine learning classification algorithm is proposed. First, the time delay signals of circuits under different voltages are collected, and then determine whether there is a Trojan horse in the circuit depending on the time delay. A polynomial regression algorithm is combined to fit the delay data, and Trojan horse feature library can be established based on the regression function. The experimental results show that the 19 different voltages of 2 000 groups of delay units are extracted to compare. When the number of voltages considered comprehensively is 19, its prediction accuracy reaches the highest 95.2%. The proposed classification and regression algorithms can improve the recognition accuracy and automation of the hardware Trojan.
作者
宋钛
黄正峰
闫爱斌
Song Tai;Huang Zhengfeng;Yan Aibin(School of Integrated Circuits,Anhui University,Hefei 230601;School of Microelectronics,Hefei University of Technology,Hefei 230009;School of Computer Science and Technology,Anhui University,Hefei 230601)
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2022年第4期515-521,共7页
Journal of Computer-Aided Design & Computer Graphics
基金
国家自然科学基金(61874156)。
关键词
硬件木马
机器学习
干净电路
多项式回归
hardware Trojan horse
machine learning
clean circuit
polynomial regression
