针对侧信道检测方法检出率不高的问题,提出一种基于主成分分析结合马氏距离的检测方法.通过对芯片功耗进行建模分析,首先采用主成分分析法对旁路信息中的微小差异进行放大提取,获取主特征,然后使用马氏距离进行判别区分,识别硬件木马.基于自主设计的FPGA(field-programmable gate array)检测平台进行实验验证,结果表明:采用基于主成分分析结合马氏距离的硬件木马检测方法可以有效检测出占母本电路面积0.6%左右的硬件木马. Aiming at the problem that the detection rate of side channel detection is not high, this paper proposes a detection method based on principal component analysis combined with Mahalanobis distance. By modeling the power consumption of the chip, the principal component analysis (PCA) And then use Mahalanobis distance to discriminate and distinguish the hardware Trojans.According to the field-programmable gate array (FPGA) detection platform designed by ourselves, the experimental results show that the main features are based on the combination of principal components analysis Mahalanobis distance from the hardware Trojan detection method can effectively detect the female circuit area of ​​about 0.6% of the hardware Trojans.