Extension and performance/accuracy formulation for optimal GeAr-based approximate adder designs

    Research output: Contribution to journalArticle

    Abstract

    Approximate computing is a promising solution for future energy-efficient designs because it can provide great improvements in performance, area and/or energy consumption over traditional exact-computing designs for non-critical error-tolerant applications. However, the most challenging issue in designing approximate circuits is how to guarantee the pre-specified computation accuracy while achieving energy reduction and performance improvement. To address this problem, this paper starts from the state-of-the-art general approximate adder model (GeAr) and extends it for more possible approximate design candidates by relaxing the design restrictions. And then a maximum-error-distance-based performance/accuracy formulation, which can be used to select the performance/energy-accuracy optimal design from the extended design space, is proposed. Our evaluation results show the effectiveness of the proposed method in terms of area overhead, performance, energy consumption, and computation accuracy.

    Original languageEnglish
    Pages (from-to)1014-1024
    Number of pages11
    JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
    VolumeE101A
    Issue number7
    DOIs
    Publication statusPublished - 2018 Jul 1

    Fingerprint

    Approximate Design
    Adders
    Gears
    Formulation
    Energy Consumption
    Energy utilization
    Computing
    Approximate Model
    Energy
    Energy Efficient
    Restriction
    Design
    Networks (circuits)
    Evaluation

    Keywords

    • Adder
    • Approximate computing
    • Energy-efficient
    • GeAr

    ASJC Scopus subject areas

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

    Cite this

    @article{79098d895c884ec2b230a4be502b4541,
    title = "Extension and performance/accuracy formulation for optimal GeAr-based approximate adder designs",
    abstract = "Approximate computing is a promising solution for future energy-efficient designs because it can provide great improvements in performance, area and/or energy consumption over traditional exact-computing designs for non-critical error-tolerant applications. However, the most challenging issue in designing approximate circuits is how to guarantee the pre-specified computation accuracy while achieving energy reduction and performance improvement. To address this problem, this paper starts from the state-of-the-art general approximate adder model (GeAr) and extends it for more possible approximate design candidates by relaxing the design restrictions. And then a maximum-error-distance-based performance/accuracy formulation, which can be used to select the performance/energy-accuracy optimal design from the extended design space, is proposed. Our evaluation results show the effectiveness of the proposed method in terms of area overhead, performance, energy consumption, and computation accuracy.",
    keywords = "Adder, Approximate computing, Energy-efficient, GeAr",
    author = "Ken Hayamizu and Nozomu Togawa and Masao Yanagisawa and Youhua Shi",
    year = "2018",
    month = "7",
    day = "1",
    doi = "10.1587/transfun.E101.A.1014",
    language = "English",
    volume = "E101A",
    pages = "1014--1024",
    journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
    issn = "0916-8508",
    publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
    number = "7",

    }

    TY - JOUR

    T1 - Extension and performance/accuracy formulation for optimal GeAr-based approximate adder designs

    AU - Hayamizu, Ken

    AU - Togawa, Nozomu

    AU - Yanagisawa, Masao

    AU - Shi, Youhua

    PY - 2018/7/1

    Y1 - 2018/7/1

    N2 - Approximate computing is a promising solution for future energy-efficient designs because it can provide great improvements in performance, area and/or energy consumption over traditional exact-computing designs for non-critical error-tolerant applications. However, the most challenging issue in designing approximate circuits is how to guarantee the pre-specified computation accuracy while achieving energy reduction and performance improvement. To address this problem, this paper starts from the state-of-the-art general approximate adder model (GeAr) and extends it for more possible approximate design candidates by relaxing the design restrictions. And then a maximum-error-distance-based performance/accuracy formulation, which can be used to select the performance/energy-accuracy optimal design from the extended design space, is proposed. Our evaluation results show the effectiveness of the proposed method in terms of area overhead, performance, energy consumption, and computation accuracy.

    AB - Approximate computing is a promising solution for future energy-efficient designs because it can provide great improvements in performance, area and/or energy consumption over traditional exact-computing designs for non-critical error-tolerant applications. However, the most challenging issue in designing approximate circuits is how to guarantee the pre-specified computation accuracy while achieving energy reduction and performance improvement. To address this problem, this paper starts from the state-of-the-art general approximate adder model (GeAr) and extends it for more possible approximate design candidates by relaxing the design restrictions. And then a maximum-error-distance-based performance/accuracy formulation, which can be used to select the performance/energy-accuracy optimal design from the extended design space, is proposed. Our evaluation results show the effectiveness of the proposed method in terms of area overhead, performance, energy consumption, and computation accuracy.

    KW - Adder

    KW - Approximate computing

    KW - Energy-efficient

    KW - GeAr

    UR - http://www.scopus.com/inward/record.url?scp=85049392872&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85049392872&partnerID=8YFLogxK

    U2 - 10.1587/transfun.E101.A.1014

    DO - 10.1587/transfun.E101.A.1014

    M3 - Article

    AN - SCOPUS:85049392872

    VL - E101A

    SP - 1014

    EP - 1024

    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 - 7

    ER -