Hardware-Trojan Detection Based on the Structural Features of Trojan Circuits Using Random Forests

Tatsuki KURIHARA; Nozomu TOGAWA

doi:10.1587/transfun.2021EAP1091

Hardware-Trojan Detection Based on the Structural Features of Trojan Circuits Using Random Forests

Tatsuki KURIHARA, Nozomu TOGAWA

Full Text Views

0

Cite this

Summary :

Recently, with the spread of Internet of Things (IoT) devices, embedded hardware devices have been used in a variety of everyday electrical items. Due to the increased demand for embedded hardware devices, some of the IC design and manufacturing steps have been outsourced to third-party vendors. Since malicious third-party vendors may insert malicious circuits, called hardware Trojans, into their products, developing an effective hardware-Trojan detection method is strongly required. In this paper, we propose 25 hardware-Trojan features focusing on the structure of trigger circuits for machine-learning-based hardware-Trojan detection. Combining the proposed features into 11 existing hardware-Trojan features, we totally utilize 36 hardware-Trojan features for classification. Then we classify the nets in an unknown netlist into a set of normal nets and Trojan nets based on a random-forest classifier. The experimental results demonstrate that the average true positive rate (TPR) becomes 64.2% and the average true negative rate (TNR) becomes 100.0%. They improve the average TPR by 14.8 points while keeping the average TNR compared to existing state-of-the-art methods. In particular, the proposed method successfully finds out Trojan nets in several benchmark circuits, which are not found by the existing method.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E105-A No.7 pp.1049-1060

Publication Date: 2022/07/01

Publicized: 2022/01/07

Online ISSN: 1745-1337

DOI: 10.1587/transfun.2021EAP1091

Type of Manuscript: PAPER

Category: VLSI Design Technology and CAD

Authors

Tatsuki KURIHARA
Waseda University
Nozomu TOGAWA
Waseda University

Keyword

hardware Trojan, trigger circuit, gate-level netlist, machine learning, random forest

Cite this

Copy

Tatsuki KURIHARA, Nozomu TOGAWA, "Hardware-Trojan Detection Based on the Structural Features of Trojan Circuits Using Random Forests" in IEICE TRANSACTIONS on Fundamentals, vol. E105-A, no. 7, pp. 1049-1060, July 2022, doi: 10.1587/transfun.2021EAP1091.
Abstract: Recently, with the spread of Internet of Things (IoT) devices, embedded hardware devices have been used in a variety of everyday electrical items. Due to the increased demand for embedded hardware devices, some of the IC design and manufacturing steps have been outsourced to third-party vendors. Since malicious third-party vendors may insert malicious circuits, called hardware Trojans, into their products, developing an effective hardware-Trojan detection method is strongly required. In this paper, we propose 25 hardware-Trojan features focusing on the structure of trigger circuits for machine-learning-based hardware-Trojan detection. Combining the proposed features into 11 existing hardware-Trojan features, we totally utilize 36 hardware-Trojan features for classification. Then we classify the nets in an unknown netlist into a set of normal nets and Trojan nets based on a random-forest classifier. The experimental results demonstrate that the average true positive rate (TPR) becomes 64.2% and the average true negative rate (TNR) becomes 100.0%. They improve the average TPR by 14.8 points while keeping the average TNR compared to existing state-of-the-art methods. In particular, the proposed method successfully finds out Trojan nets in several benchmark circuits, which are not found by the existing method.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2021EAP1091/_p

Copy

@ARTICLE{e105-a_7_1049,
author={Tatsuki KURIHARA, Nozomu TOGAWA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Hardware-Trojan Detection Based on the Structural Features of Trojan Circuits Using Random Forests},
year={2022},
volume={E105-A},
number={7},
pages={1049-1060},
abstract={Recently, with the spread of Internet of Things (IoT) devices, embedded hardware devices have been used in a variety of everyday electrical items. Due to the increased demand for embedded hardware devices, some of the IC design and manufacturing steps have been outsourced to third-party vendors. Since malicious third-party vendors may insert malicious circuits, called hardware Trojans, into their products, developing an effective hardware-Trojan detection method is strongly required. In this paper, we propose 25 hardware-Trojan features focusing on the structure of trigger circuits for machine-learning-based hardware-Trojan detection. Combining the proposed features into 11 existing hardware-Trojan features, we totally utilize 36 hardware-Trojan features for classification. Then we classify the nets in an unknown netlist into a set of normal nets and Trojan nets based on a random-forest classifier. The experimental results demonstrate that the average true positive rate (TPR) becomes 64.2% and the average true negative rate (TNR) becomes 100.0%. They improve the average TPR by 14.8 points while keeping the average TNR compared to existing state-of-the-art methods. In particular, the proposed method successfully finds out Trojan nets in several benchmark circuits, which are not found by the existing method.},
keywords={},
doi={10.1587/transfun.2021EAP1091},
ISSN={1745-1337},
month={July},}

Copy

TY - JOUR
TI - Hardware-Trojan Detection Based on the Structural Features of Trojan Circuits Using Random Forests
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1049
EP - 1060
AU - Tatsuki KURIHARA
AU - Nozomu TOGAWA
PY - 2022
DO - 10.1587/transfun.2021EAP1091
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E105-A
IS - 7
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - July 2022
AB - Recently, with the spread of Internet of Things (IoT) devices, embedded hardware devices have been used in a variety of everyday electrical items. Due to the increased demand for embedded hardware devices, some of the IC design and manufacturing steps have been outsourced to third-party vendors. Since malicious third-party vendors may insert malicious circuits, called hardware Trojans, into their products, developing an effective hardware-Trojan detection method is strongly required. In this paper, we propose 25 hardware-Trojan features focusing on the structure of trigger circuits for machine-learning-based hardware-Trojan detection. Combining the proposed features into 11 existing hardware-Trojan features, we totally utilize 36 hardware-Trojan features for classification. Then we classify the nets in an unknown netlist into a set of normal nets and Trojan nets based on a random-forest classifier. The experimental results demonstrate that the average true positive rate (TPR) becomes 64.2% and the average true negative rate (TNR) becomes 100.0%. They improve the average TPR by 14.8 points while keeping the average TNR compared to existing state-of-the-art methods. In particular, the proposed method successfully finds out Trojan nets in several benchmark circuits, which are not found by the existing method.
ER -