面向硬件木马检测的旁路信号特征选择方法
发布时间：2018-09-03 11:55
【摘要】：针对基于旁路分析的硬件木马检测中存在的旁路信号冗余以及高维问题,探究特征选择方法在去除冗余、降低旁路信号维数方面的可行性,提出了一种以类内类间距离作为可分性判据的特征选择方法对旁路信号进行预先处理。首先分析了IC芯片旁路信号的特征选择问题,然后阐述了基于类内类间距离的可分性判据以及特征选择搜索算法,最后在FPGA密码芯片中植入硬件木马,并基于K-L方法进行检测实验,通过对旁路信号进行特征选择前后的木马检测效果对比发现,该特征选择方法能有助于分辨出无木马的"金片"与含木马芯片之间旁路信号的统计特征差异,更好地实现硬件木马的检测。
[Abstract]:Aiming at the redundancy and high dimension of bypass signal in the detection of hardware Trojan horse based on bypass analysis, the feasibility of feature selection in removing redundancy and reducing the dimension of bypass signal is explored. In this paper, a method of feature selection based on the inter-class distance as the criterion of separability is proposed to preprocess the by-pass signal. In this paper, the feature selection problem of bypass signal in IC chip is analyzed, then the separability criterion based on inter-class distance and the feature selection search algorithm are described. Finally, a hardware Trojan is implanted in the FPGA cipher chip. And based on K-L method, the detection results of Trojan horse before and after feature selection of bypass signal are compared and found. The method of feature selection can help to distinguish the statistical characteristics of the bypass signal between the trojaned "gold chip" and the trojaned chip, and to better realize the detection of the hardware Trojan horse.
【作者单位】： 军械工程学院信息工程系;
【基金】：国家自然科学基金(No.61271152,No.51377170)
【分类号】：TN407;TP309
本文编号：2219906
[Abstract]:Aiming at the redundancy and high dimension of bypass signal in the detection of hardware Trojan horse based on bypass analysis, the feasibility of feature selection in removing redundancy and reducing the dimension of bypass signal is explored. In this paper, a method of feature selection based on the inter-class distance as the criterion of separability is proposed to preprocess the by-pass signal. In this paper, the feature selection problem of bypass signal in IC chip is analyzed, then the separability criterion based on inter-class distance and the feature selection search algorithm are described. Finally, a hardware Trojan is implanted in the FPGA cipher chip. And based on K-L method, the detection results of Trojan horse before and after feature selection of bypass signal are compared and found. The method of feature selection can help to distinguish the statistical characteristics of the bypass signal between the trojaned "gold chip" and the trojaned chip, and to better realize the detection of the hardware Trojan horse.
【作者单位】： 军械工程学院信息工程系;
【基金】：国家自然科学基金(No.61271152,No.51377170)
【分类号】：TN407;TP309
【相似文献】
相关期刊论文 前1条
1 王新峰,邱静,刘冠军;机电BIT故障特征选择方法研究[J];中国机械工程;2004年12期
，本文编号：2219906
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/2219906.html
