陆军工程大学石家庄校区 装备模拟训练中心, 石家庄 050003
出版日期:2019-10-28发布日期:2019-11-01通讯作者:李雄伟，男，副教授，博士生导师，电话(Tel.)：0311-87994929；E-mail：lxw-wys@163.com.作者简介:王晓晗(1992-)，男，河北省衡水市人，博士生，主要研究方向为芯片安全.基金资助:国家自然科学基金资助项目(61602505)Test Pattern Generation Method for Hardware Trojan Detection Based on Artificial Bee Colony
WANG Xiaohan,WANG Tao,LI Xiongwei,ZHANG Yang,HUANG ChangyangEquipment Simulation Training Center, Army Engineering University Shijiazhuang Campus, Shijiazhuang 050003, China
Online:2019-10-28Published:2019-11-01摘要/Abstract
摘要： 针对已有测试向量生成方法对以电路惰性节点作为输入的硬件木马触发覆盖率低的问题，提出了一种基于人工蜂群的测试向量生成方法.首先分析了用于触发惰性节点组合的测试向量的分布规律，并构建数学模型对其进行描述；然后利用人工蜂群算法生成测试向量，结合其分布规律对局部区域进行高效搜索以发现能触发更多惰性节点组合的测试向量，同时对全局进行快速搜索，有效避免了“早期收敛”问题.实验结果表明：使用本文方法生成的测试向量测试电路，对电路中惰性节点组合的平均触发覆盖率达到95.86%，与已有方法相比提高了22.43%，具有更好的硬件木马激活效果.
关键词: 硬件木马； 集成电路； 惰性节点； 人工蜂群算法； 激活策略
Abstract: The exiting test pattern generation method have the problem of low trigger coverage for hardware Trojan detection. In order to solve this problem, a test pattern generation method based on artificial bee colony algorithm is proposed. Firstly, the distribution regularity of test patterns which can trigger the combination of inactive nets is analyzed. And the mathematical model is constructed to describe the test pattern. Then, the test pattern is generated by artificial bee colony algorithm. Combining with its distribution regularity, this method can search local regions efficiently to find test patterns that can trigger more combinations of inactive nets. At the same time, it can search global world quickly and effectively avoid the problem of "premature convergence". The experimental results show that using the test vectors generated by this method to test circuit, the average trigger coverage rate of hardware Trojan can reach 95.86%. Compared with the existing method, this method improves 22.43%, and has better hardware Trojan activation effect.
Key words: hardware Trojan; integrated circuit (IC); inactive net; artificial bee colony algorithm; activation strategy
PDF全文下载地址:
点我下载PDF
