一种深度学习的硬件木马检测算法

doi:10.19665/j.issn1001-2400.2019.06.006

西安电子科技大学学报 ›› 2019, Vol. 46 ›› Issue (6): 37-45.doi: 10.19665/j.issn1001-2400.2019.06.006

一种深度学习的硬件木马检测算法

刘志强^1,²,张铭津^1,²(),池源³,李云松¹

^1. 西安电子科技大学综合业务网理论及关键技术国家重点实验室,陕西西安 710071
^2. 中国科学院光谱成像技术重点实验室,陕西西安,710119
^3. 电子元器件可靠性物理及其应用技术重点实验室,广东广州,510610

收稿日期:2019-05-05 出版日期:2019-12-20 发布日期:2019-12-21
通讯作者: 张铭津
作者简介:刘志强(1995—),男,西安电子科技大学硕士研究生, E-mail:1973468275@qq.com
基金资助:
国家自然科学基金(61902293);陕西省自然科学基础研究计划(2018JQ6028);中央高校基本科研业务费专项基金(XJS17109);中央高校基本科研业务费专项基金(JBX180102);中国博士后科学基金面上项目(2017M623125);电子元器件可靠性物理及其应用技术重点实验室开放基金(17D03-ZHD201701);国家部委纵向“十三五”公用信息系统装备研究项目(3151****403);中国科学院光谱成像技术重点实验室开放基金(LSIT201901W)

Hardware Trojan detection algorithm based on deep learning

LIU Zhiqiang^1,²,ZHANG Mingjin^1,²(),CHI Yuan³,LI Yunsong¹

^1. State Key Laboratory of Integrated Service Networks, Xidian University, Xi’an 710071, China
^2. CAS Key Laboratory of Spectral Imaging Technology, Xi’an 710119, China
^3. Science and Technology on Reliability Physics and Application of Electronic Component Laboratory, the Fifth Electronics Research Institute of Ministry of Industry and Information Technology,Guangzhou 510610, China

Received:2019-05-05 Online:2019-12-20 Published:2019-12-21
Contact: Mingjin ZHANG

摘要/Abstract

摘要：

由于传统的硬件木马检测均采用功能测试等电信号检测技术,检测方法存在成本高、漏检率高和效率低下等问题,对此提出了一种深度学习的非电信号硬件木马检测算法。该算法首先利用增强残差网络将低分辨率芯片显微图像转换为高分辨率芯片显微图像; 然后通过循环一致对抗生成网络将该高分辨率图像生成与母版图像同源的芯片显微图像, 生成的芯片显微图像通过二阶微分图像增强算法区分出目标区域与背景区域,并结合阈值分割算法将目标区域分割出来; 最后通过数学形态学操作去除由于工业噪声产生的微小干扰,利用变化检测算法检测芯片中存在的硬件木马。通过在芯片显微图像数据集上的实验显示,基于深度学习的硬件木马检测方法正检率高达约92.4%,与传统的电信号检测方法相比,精度更高,速度更快,且操作更简易。

关键词: 硬件木马检测, 深度学习, 图像增强, 图像分割, 数学形态学操作, 变化检测

Abstract:

The traditional way of hardware trojan detection based on electrical signal detection has problems of low positive rate, low efficiency and high cost. To solve these problems, we propose a new way of hardware trojan detection based on deep learning which is not electrical signal detection. First, the algorithm changes chip microscopic images of low resolution into chip microscopic images of high resolution by using an enhanced residual network. Then these chip microscopic images of high resolution will generate another chip microscopic images which are similar to those of the golden model. The algorithm for image enhancement distinguishes between target area and background area by combining with the algorithm of image segmentation. Finally, we use the change detection algorithm to detect the hardware trojans existing in the chip after removing minor interference due to industrial noise. Through the experiments on the micrograph dataset of the chip, the positive detection rate of the hardware trojan detection method based on deep learning is as high as 92.4%. Compared with the traditional electrical signal detection method, our algorithm has the advantages of higher precision, faster speed, and easier operation.

Key words: hardware trojan detection, deep learning, image enhancement, image segmentation, mathematical morphology, change detection

中图分类号:

TP274

刘志强,张铭津,池源,李云松. 一种深度学习的硬件木马检测算法[J]. 西安电子科技大学学报, 2019, 46(6): 37-45.

LIU Zhiqiang,ZHANG Mingjin,CHI Yuan,LI Yunsong. Hardware Trojan detection algorithm based on deep learning[J]. Journal of Xidian University, 2019, 46(6): 37-45.

图/表 6

图1

图2

图3

图4

表1

图5

参考文献 27

[1]	YANG K, HICKS M, DONG Q , et al. A2: Analog Malicious Hardware[C]//Proceedings of the 2016 IEEE Symposium on Security and Privacy. Piscataway: IEEE, 2016: 18-37.
[2]	BAO C, FORTE D, SRIVASTAVA A , et al. Reverse Engineering-based Hardware Trojan Detection[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2016,35(1):49-57. doi: 10.1109/TCAD.2015.2488495
[3]	薛明富, 王箭, 胡爱群 . 自适应优化的二元分类型硬件木马检测方法[J]. 计算机学报, 2018,41(2):439-451.
	XUE Mingfu WANG Jian, HU Aiqun . Adaptive Optimization of Two-class Classification-based Hardware Trojan Detection Method[J]. Chinese Journal of Computers, 2018,41(2):439-451.
[4]	NOURIAN M A, FAZELI M, HELY D . Hardware Trojan Detection Using an Advised Genetic Algorithm Based Logic Testing[J]. Journal of Electronic Testing, 2018,34(4):461-470. doi: 10.1007/s10836-018-5739-4
[5]	ZHANG Y, QUAN H, LI X , et al. Golden-free Processor Hardware Trojan Detection Using Bit Power Consistency Analysis[J]. Journal of Electronic Testing, 2018,34(3):305-312. doi: 10.1007/s10836-018-5715-z
[6]	DE SOUZA FARIA G, KIM H Y . Differential Audio Analysis: a New Side-channel Attack on PIN Pads[J]. International Journal of Information Security, 2019,18(1):73-84. doi: 10.1007/s10207-018-0403-7
[7]	杨亚君, 陈章 . 分块制造下硬件木马攻击方法及安全性分析[J]. 西安电子科技大学学报, 2019,46(4):167-175.
	YANG Yajun, CHEN Zhang . Hardware Trojan Attack Methods and Security Analysis under Split Manufacturing[J]. Journal of Xidian University, 2019,46(4):167-175.
[8]	赵毅强, 解啸天, 刘燕江 , 等. 基于电路活性测度的硬件木马检测方法[J]. 华中科技大学学报, 2018,46(2):90-94.
	ZHAO Yiqiang, XIE Xiaotian, LIU Yanjiang , et al. Hardware Trojan Detection Method Based on Circuit Activity[J]. Journal of Huazhong University of Science and Technology, 2018,46(2):90-94.
[9]	NEJAT A, HELY D, BEROULLE V . ESCALATION: Leveraging Logic Masking to Facilitate Path-delay-based Hardware Trojan Detection Methods[J]. Journal of Hardware and Systems Security, 2018,2(1):83-96. doi: 10.1007/s41635-018-0033-6
[10]	BAO C, FORTE D, SRIVASTAVA A . On Reverse Engineering-based Hardware Trojan Detection[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2016,35(1):49-57. doi: 10.1109/TCAD.2015.2488495
[11]	RESHMA K, PRIYATHARISHINI M, NIRMALA D M . Hardware Trojan Detection Using Deep Learning Technique[C]//Advances in Intelligent Systems and Computing: 898. Heidelberg: Springer Verlag, 2019: 671-680.
[12]	MENDIS G J, WEI J, MADANAYAKE A . Deep Learning based Radio-signal Identification with Hardware Design[J]. IEEE Transactions on Aerospace and Electronic Systems, 2019, DOI: 10.1109/TAES.2019.2891155. doi: 10.1109/TAES.2018.2828218 pmid: 31823972
[13]	ABBASI H, EZZATI-JIVAN N, BELLAICHE M , et al. Machine Learning-based EDoS Attack Detection Technique Using Execution Trace Analysis[J]. Journal of Hardware and Systems Security, 2019,3(2):164-176. doi: 10.1007/s41635-018-0061-2
[14]	王世平, 毕笃彦, 刘坤 , 等. 一种多映射卷积神经网络的超分辨率重建算法[J]. 西安电子科技大学学报, 2018,45(4):155-160.
	WANG Shiping, BI Duyan, LIU Kun , et al. Multi-mapping Convolution Neural Network for the Image Super-resolution Algorithm[J]. Journal of Xidian University, 2018,45(4):155-160.
[15]	陈晓范, 申海杰, 边倩 , 等. 结合注意力机制的人脸超分辨率重建[J]. 西安电子科技大学学报, 2019,46(3):148-153.
	CHEN Xiaofan, SHEN Haijie, BIAN Qian , et al. Face Image Super-resolution with an Attention Mechanism[J]. Journal of Xidian University, 2019,46(3):148-153.
[16]	ZHANG M J, WANG N N, LI Y S , et al. Bionic Face Sketch Generator[J]. IEEE Transactions on Cybernetics, 2019, DOI: 10.1109/TCYB.2019.2924589. doi: 10.1109/TCYB.2019.2956316 pmid: 31869813
[17]	ZHANG M, WANG R, GAO X , et al. Dual-transfer Face Sketch-Photo Synjournal[J]. IEEE Transactions on Image Processing, 2019,28(2):642-657. doi: 10.1109/TIP.2018.2869688 pmid: 30222563
[18]	ZHANG M, LI J, WANG N , et al. Recognition of Facial Sketch Styles[J]. Neurocomputing, 2015,149(PC):1188-1197. doi: 10.1109/TIP.2018.2869688 pmid: 30222563
[19]	DONG C, LOY C C, TANG X , et al. Learning a Deep Convolutional Network for Image Super-resolution[C]// Lecture Notes in Computer Science: 8692. Heidelberg: Springer Verlag, 2014: 184-199.
[20]	KIM J, LEE J K, LEE K M . Accurate Image Super-resolution Using Very Deep Convolutional Networks[C]// Proceedings of the 2016 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington: IEEE Computer Society, 2016: 1646-1654.
[21]	DONG C, LOY C C, TANG X . Accelerating the Super-resolution Convolutional Neural Network[C]//Lecture Notes in Computer Science: 9906. Heidelberg: Springer Verlag, 2016: 391-407.
[22]	LIM B, SON S, KIM H , et al. Enhanced Deep Residual Networks for Single Image Super-resolution[C]// Proceedings of the 2017 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Washington: IEEE Computer Society, 2017: 1132-1140.
[23]	MOSSER L, DUBRULE O, BLUNT M J . Stochastic Reconstruction of an Oolitic Limestone by Generative Adversarial Networks[J]. Transport in Porous Media, 2018,125(1):81-103. doi: 10.1007/s11242-018-1039-9
[24]	KAWAI Y, SEO M, CHEN Y W . Automatic Generation of Facial Expression Using Generative Adversarial Nets[C]//Proceedings of the 2018 IEEE 7th Global Conference on Consumer Electronics. Piscataway: IEEE, 2018: 329-330.
[25]	ZHU J Y, PARK T, ISOLA P , et al. Unpaired Image-to-image Translation Using Cycle-consistent Adversarial Networks[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 2242-2251.
[26]	ZHANG M J, WANG N N, LI Y S , et al. Deep Latent Low-rank Representation for Face Sketch Synjournal[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, DOI: 10.1109/TNNLS.2018.2890017. doi: 10.1109/TNNLS.2019.2955438 pmid: 31869807
[27]	ZHANG M, LI J, WANG N , et al. Compositional Model-based Sketch Generator in Facial Entertainment[J]. IEEE Transactions on Cybernetics, 2018,48(3):904-915. doi: 10.1109/TCYB.2017.2664499 pmid: 28212105

模块	ta / TA (non)	td / TD (non)	ta / TA (sr)	td / TD (sr)	ta / TA (ge)	td / TD (ge)	ta / TA (sr+ge)	td / TD (sr+ge)
block1	23/28 (0.821)	20/29 (0.689)	25/28 (0.892)	25/29 (0.862)	24/28 (0.857)	23/29 (0.793)	26/28 (0.928)	26/29 (0.896)
block2	76/85 (0.894)	52/64 (0.812)	80/85 (0.941)	58/64 (0.906)	78/85 (0.917)	54/64 (0.843)	80/85 (0.898)	60/64 (0.937)
block3	38/43 (0.883)	33/41 (0.804)	40/43 (0.930)	38/41 (0.926)	38/43 (0.883)	34/41 (0.829)	40/43 (0.930)	39/41 (0.951)
block4	25/33 (0.757)	20/26 (0.769)	28/33 (0.848)	24/26 (0.923)	26/33 (0.787)	22/26 (0.846)	29/33 (0.878)	26/26 (1.000)
block5	14/19 (0.736)	15/24 (0.625)	16/19 (0.842)	21/24 (0.875)	14/19 (0.736)	16/24 (0.666)	17/19(0.894)	22/24 (0.916)
block6	12/19 (0.631	14/20 (0.700)	15/19 (0.789)	18/20 (0.900)	13/19 (0,684)	16/20 (0.800)	16/19 (0.842)	19/20 (0.950)
block7	44/59 (0.745)	83/94 (0.882)	51/59 (0.864)	89/94 (0.893)	48/59 (0.813)	86/94 (0.914)	52/59 (0.881)	91/94 (0.968)
block8	35/41 (0.853)	11/16 (0.687)	38/41 (0.926)	13/16 (0.812)	35/41 (0.853)	11/16 (0.687)	38/41 (0.926)	14/16 (0.875)
block9	23/25 (0.920)	26/35 (0.742)	24/25 (0,960)	31/35 (0.885)	23/25 (0.920)	28/35 (0.800)	24/25 (0.960)	32/35 (0.914)
block10	18/23 (0.695)	15/25 (0.600)	21/23 (0.913)	18/25 (0.720)	20/23 (0.869)	16/25 (0.640)	21/23 (0.913)	21/25 (0.840)
Average	(0.793)	(0.731)	(0.890)	(0.870)	(0.831)	(0.781)	(0.905)	(0.924)

一种深度学习的硬件木马检测算法

Hardware Trojan detection algorithm based on deep learning

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 27

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	刘佳玮,张文辉,寇晓丽,李雁妮. 增强型深度对抗样本攻击防御算法[J]. 西安电子科技大学学报, 2021, 48(6): 23-31.
[2]	李鹏,冯存前,许旭光,唐子翔. 一种利用贝叶斯优化的弹道目标微动分类网络[J]. 西安电子科技大学学报, 2021, 48(5): 139-148.
[3]	孙彦景,魏力,张年龙,云霄,董锴文,葛敏,程小舟,侯晓峰. 联合DD-GAN和全局特征的井下人员重识别方法[J]. 西安电子科技大学学报, 2021, 48(5): 201-211.
[4]	闫佳,曹玉东,任佳兴,陈冬昊,李晓会. 深度非对称压缩型哈希算法[J]. 西安电子科技大学学报, 2021, 48(5): 212-221.
[5]	宁阳,杜建超,韩硕,杨传凯. 改进DeeplabV3+的火焰分割与火情分析方法[J]. 西安电子科技大学学报, 2021, 48(5): 38-46.
[6]	周鹏,杨军. 采用神经网络架构搜索的遥感影像分割方法[J]. 西安电子科技大学学报, 2021, 48(5): 47-57.
[7]	戚艳军,孔月萍,王佳婧,朱旭东. 一种LSTM与CNN相结合的步态识别方法[J]. 西安电子科技大学学报, 2021, 48(5): 78-85.
[8]	张宇浩,程培涛,张书豪,王秀美. 一种自适应权重学习的轻量超分辨率重建网络[J]. 西安电子科技大学学报, 2021, 48(5): 15-22.
[9]	宋建锋,苗启广,王崇晓,徐浩,杨瑾. 注意力机制的多尺度单目标跟踪算法[J]. 西安电子科技大学学报, 2021, 48(5): 110-116.
[10]	回海生,张雪英,吴泽林,李凤莲. 一种主辅路径注意力补偿的脑卒中病灶分割方法[J]. 西安电子科技大学学报, 2021, 48(4): 200-208.
[11]	宋剑桥,王峰,牛锦,师泽洲,马军辉. 一种面向时空神经网络的潜在情绪识别方法[J]. 西安电子科技大学学报, 2021, 48(4): 159-167.
[12]	孙豪杰,李苗钰,章盼盼,许鹏飞. 用于面瘫分级的自监督非对称特征学习方法[J]. 西安电子科技大学学报, 2021, 48(3): 115-122.
[13]	王俊祥,黄霖,张影,倪江群,林朗. 低复杂度的增强图像来源检测算法[J]. 西安电子科技大学学报, 2021, 48(1): 96-106.
[14]	张华,高浩然,杨兴国,李文敏,高飞,温巧燕. TargetedFool:一种实现有目标攻击的算法[J]. 西安电子科技大学学报, 2021, 48(1): 149-159.
[15]	杨宏宇,曾仁韵. 一种深度学习的网络安全态势评估方法[J]. 西安电子科技大学学报, 2021, 48(1): 183-190.