Register-transfer-level (RTL) information is hardly available for hardware Trojan detection. In this paper, four RTL Trojan features related to branching statement are proposed. The Minimum Redundancy Maximum Relevance (mRMR) feature selection is applied to the proposed Trojan features to determine the recommended feature combinations. The feature combinations are then tested using different machine learning concepts in order to determine the best approach for classifying Trojan and normal branches. The result shows that a Decision Tree classification algorithm with all the four proposed Trojan features can achieve an average true positive detection rate of 93.72% on unseen test data.
Hau Sim CHOO
Universiti Teknologi Malaysia
Chia Yee OOI
Universiti Teknologi Malaysia
Michiko INOUE
Nara Institute of Science and Technology (NAIST)
Nordinah ISMAIL
Universiti Teknologi Malaysia
Mehrdad MOGHBEL
Universiti Teknologi Malaysia
Chee Hoo KOK
Universiti Teknologi Malaysia
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hau Sim CHOO, Chia Yee OOI, Michiko INOUE, Nordinah ISMAIL, Mehrdad MOGHBEL, Chee Hoo KOK, "Register-Transfer-Level Features for Machine-Learning-Based Hardware Trojan Detection" in IEICE TRANSACTIONS on Fundamentals,
vol. E103-A, no. 2, pp. 502-509, February 2020, doi: 10.1587/transfun.2019EAP1044.
Abstract: Register-transfer-level (RTL) information is hardly available for hardware Trojan detection. In this paper, four RTL Trojan features related to branching statement are proposed. The Minimum Redundancy Maximum Relevance (mRMR) feature selection is applied to the proposed Trojan features to determine the recommended feature combinations. The feature combinations are then tested using different machine learning concepts in order to determine the best approach for classifying Trojan and normal branches. The result shows that a Decision Tree classification algorithm with all the four proposed Trojan features can achieve an average true positive detection rate of 93.72% on unseen test data.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2019EAP1044/_p
Copy
@ARTICLE{e103-a_2_502,
author={Hau Sim CHOO, Chia Yee OOI, Michiko INOUE, Nordinah ISMAIL, Mehrdad MOGHBEL, Chee Hoo KOK, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Register-Transfer-Level Features for Machine-Learning-Based Hardware Trojan Detection},
year={2020},
volume={E103-A},
number={2},
pages={502-509},
abstract={Register-transfer-level (RTL) information is hardly available for hardware Trojan detection. In this paper, four RTL Trojan features related to branching statement are proposed. The Minimum Redundancy Maximum Relevance (mRMR) feature selection is applied to the proposed Trojan features to determine the recommended feature combinations. The feature combinations are then tested using different machine learning concepts in order to determine the best approach for classifying Trojan and normal branches. The result shows that a Decision Tree classification algorithm with all the four proposed Trojan features can achieve an average true positive detection rate of 93.72% on unseen test data.},
keywords={},
doi={10.1587/transfun.2019EAP1044},
ISSN={1745-1337},
month={February},}
Copy
TY - JOUR
TI - Register-Transfer-Level Features for Machine-Learning-Based Hardware Trojan Detection
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 502
EP - 509
AU - Hau Sim CHOO
AU - Chia Yee OOI
AU - Michiko INOUE
AU - Nordinah ISMAIL
AU - Mehrdad MOGHBEL
AU - Chee Hoo KOK
PY - 2020
DO - 10.1587/transfun.2019EAP1044
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E103-A
IS - 2
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - February 2020
AB - Register-transfer-level (RTL) information is hardly available for hardware Trojan detection. In this paper, four RTL Trojan features related to branching statement are proposed. The Minimum Redundancy Maximum Relevance (mRMR) feature selection is applied to the proposed Trojan features to determine the recommended feature combinations. The feature combinations are then tested using different machine learning concepts in order to determine the best approach for classifying Trojan and normal branches. The result shows that a Decision Tree classification algorithm with all the four proposed Trojan features can achieve an average true positive detection rate of 93.72% on unseen test data.
ER -