Empirical evaluation and optimization of hardware-Trojan classification for gate-level netlists based on multi-layer neural networks

    Research output: Contribution to journalArticle

    Abstract

    Recently, it has been reported that malicious third-party IC vendors often insert hardware Trojans into their products. Especially in IC design step, malicious third-party vendors can easily insert hardware Trojans in their products and thus we have to detect them efficiently. In this paper, we propose a machine-learning-based hardware-Trojan detection method for gate-level netlists using multi-layer neural networks. First, we extract 11 Trojan-net feature values for each net in a netlist. After that, we classify the nets in an unknown netlist into a set of Trojan nets and that of normal nets using multi-layer neural networks. By experimentally optimizing the structure of multi-layer neural networks, we can obtain an average of 84.8% true positive rate and an average of 70.1% true negative rate while we can obtain 100% true positive rate in some of the benchmarks, which outperforms the existing methods in most of the cases.

    Original languageEnglish
    Pages (from-to)2320-2326
    Number of pages7
    JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
    VolumeE101A
    Issue number12
    DOIs
    Publication statusPublished - 2018 Dec 1

    Fingerprint

    Multilayer Neural Network
    Multilayer neural networks
    Hardware
    Optimization
    Evaluation
    Learning systems
    Machine Learning
    Classify
    Hardware security
    Benchmark
    Unknown

    Keywords

    • Detection
    • Gate-level netlist
    • Hardware Trojan
    • Machine learning
    • Multi-layer neural networks

    ASJC Scopus subject areas

    • Signal Processing
    • Computer Graphics and Computer-Aided Design
    • Electrical and Electronic Engineering
    • Applied Mathematics

    Cite this

    @article{f669182380704c559c62d241b3e94681,
    title = "Empirical evaluation and optimization of hardware-Trojan classification for gate-level netlists based on multi-layer neural networks",
    abstract = "Recently, it has been reported that malicious third-party IC vendors often insert hardware Trojans into their products. Especially in IC design step, malicious third-party vendors can easily insert hardware Trojans in their products and thus we have to detect them efficiently. In this paper, we propose a machine-learning-based hardware-Trojan detection method for gate-level netlists using multi-layer neural networks. First, we extract 11 Trojan-net feature values for each net in a netlist. After that, we classify the nets in an unknown netlist into a set of Trojan nets and that of normal nets using multi-layer neural networks. By experimentally optimizing the structure of multi-layer neural networks, we can obtain an average of 84.8{\%} true positive rate and an average of 70.1{\%} true negative rate while we can obtain 100{\%} true positive rate in some of the benchmarks, which outperforms the existing methods in most of the cases.",
    keywords = "Detection, Gate-level netlist, Hardware Trojan, Machine learning, Multi-layer neural networks",
    author = "Kento Hasegawa and Masao Yanagisawa and Nozomu Togawa",
    year = "2018",
    month = "12",
    day = "1",
    doi = "10.1587/transfun.E101.A.2320",
    language = "English",
    volume = "E101A",
    pages = "2320--2326",
    journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
    issn = "0916-8508",
    publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
    number = "12",

    }

    TY - JOUR

    T1 - Empirical evaluation and optimization of hardware-Trojan classification for gate-level netlists based on multi-layer neural networks

    AU - Hasegawa, Kento

    AU - Yanagisawa, Masao

    AU - Togawa, Nozomu

    PY - 2018/12/1

    Y1 - 2018/12/1

    N2 - Recently, it has been reported that malicious third-party IC vendors often insert hardware Trojans into their products. Especially in IC design step, malicious third-party vendors can easily insert hardware Trojans in their products and thus we have to detect them efficiently. In this paper, we propose a machine-learning-based hardware-Trojan detection method for gate-level netlists using multi-layer neural networks. First, we extract 11 Trojan-net feature values for each net in a netlist. After that, we classify the nets in an unknown netlist into a set of Trojan nets and that of normal nets using multi-layer neural networks. By experimentally optimizing the structure of multi-layer neural networks, we can obtain an average of 84.8% true positive rate and an average of 70.1% true negative rate while we can obtain 100% true positive rate in some of the benchmarks, which outperforms the existing methods in most of the cases.

    AB - Recently, it has been reported that malicious third-party IC vendors often insert hardware Trojans into their products. Especially in IC design step, malicious third-party vendors can easily insert hardware Trojans in their products and thus we have to detect them efficiently. In this paper, we propose a machine-learning-based hardware-Trojan detection method for gate-level netlists using multi-layer neural networks. First, we extract 11 Trojan-net feature values for each net in a netlist. After that, we classify the nets in an unknown netlist into a set of Trojan nets and that of normal nets using multi-layer neural networks. By experimentally optimizing the structure of multi-layer neural networks, we can obtain an average of 84.8% true positive rate and an average of 70.1% true negative rate while we can obtain 100% true positive rate in some of the benchmarks, which outperforms the existing methods in most of the cases.

    KW - Detection

    KW - Gate-level netlist

    KW - Hardware Trojan

    KW - Machine learning

    KW - Multi-layer neural networks

    UR - http://www.scopus.com/inward/record.url?scp=85057466201&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85057466201&partnerID=8YFLogxK

    U2 - 10.1587/transfun.E101.A.2320

    DO - 10.1587/transfun.E101.A.2320

    M3 - Article

    AN - SCOPUS:85057466201

    VL - E101A

    SP - 2320

    EP - 2326

    JO - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

    JF - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

    SN - 0916-8508

    IS - 12

    ER -