Underdetermined source separation with simultaneous DOA estimation without initial value dependency

Tomoya Tachikawa, Kohei Yatabe, Yasuhiro Oikawa

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    1 Citation (Scopus)

    Abstract

    In this paper, a sparsity-based method for solving an underde-termind source separation problem is proposed. The proposed method is formulated as a convex optimization problem with two kinds of sparsity priors: sparsity in time-frequency domain and direction-of-arrival (DOA). These priors enable simultaneous estimation of DOA and sound sources, while the estimation result does not depend on an initialization method thanks to the convexity. Experiments using 4 sound sources recorded by 2 microphones confirmed that every random initial value in the proposed method resulted in the same performance which was better than the conventional methods.

    Original languageEnglish
    Title of host publication16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages161-165
    Number of pages5
    ISBN (Electronic)9781538681510
    DOIs
    Publication statusPublished - 2018 Nov 2
    Event16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Tokyo, Japan
    Duration: 2018 Sep 172018 Sep 20

    Other

    Other16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
    CountryJapan
    CityTokyo
    Period18/9/1718/9/20

    Fingerprint

    Source separation
    Direction of arrival
    arrivals
    Acoustic waves
    Convex optimization
    Microphones
    convexity
    acoustics
    microphones
    optimization
    Experiments

    Keywords

    • 1-norm
    • 2-norm
    • Convex optimization
    • Disjointness
    • Group sparsity
    • L1
    • L2
    • Primal-dual splitting

    ASJC Scopus subject areas

    • Signal Processing
    • Acoustics and Ultrasonics

    Cite this

    Tachikawa, T., Yatabe, K., & Oikawa, Y. (2018). Underdetermined source separation with simultaneous DOA estimation without initial value dependency. In 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings (pp. 161-165). [8521359] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IWAENC.2018.8521359

    Underdetermined source separation with simultaneous DOA estimation without initial value dependency. / Tachikawa, Tomoya; Yatabe, Kohei; Oikawa, Yasuhiro.

    16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 161-165 8521359.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Tachikawa, T, Yatabe, K & Oikawa, Y 2018, Underdetermined source separation with simultaneous DOA estimation without initial value dependency. in 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings., 8521359, Institute of Electrical and Electronics Engineers Inc., pp. 161-165, 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018, Tokyo, Japan, 18/9/17. https://doi.org/10.1109/IWAENC.2018.8521359
    Tachikawa T, Yatabe K, Oikawa Y. Underdetermined source separation with simultaneous DOA estimation without initial value dependency. In 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 161-165. 8521359 https://doi.org/10.1109/IWAENC.2018.8521359
    Tachikawa, Tomoya ; Yatabe, Kohei ; Oikawa, Yasuhiro. / Underdetermined source separation with simultaneous DOA estimation without initial value dependency. 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 161-165
    @inproceedings{f2d0922ad50a46daa12ac50aab15cc90,
    title = "Underdetermined source separation with simultaneous DOA estimation without initial value dependency",
    abstract = "In this paper, a sparsity-based method for solving an underde-termind source separation problem is proposed. The proposed method is formulated as a convex optimization problem with two kinds of sparsity priors: sparsity in time-frequency domain and direction-of-arrival (DOA). These priors enable simultaneous estimation of DOA and sound sources, while the estimation result does not depend on an initialization method thanks to the convexity. Experiments using 4 sound sources recorded by 2 microphones confirmed that every random initial value in the proposed method resulted in the same performance which was better than the conventional methods.",
    keywords = "1-norm, 2-norm, Convex optimization, Disjointness, Group sparsity, L1, L2, Primal-dual splitting",
    author = "Tomoya Tachikawa and Kohei Yatabe and Yasuhiro Oikawa",
    year = "2018",
    month = "11",
    day = "2",
    doi = "10.1109/IWAENC.2018.8521359",
    language = "English",
    pages = "161--165",
    booktitle = "16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - GEN

    T1 - Underdetermined source separation with simultaneous DOA estimation without initial value dependency

    AU - Tachikawa, Tomoya

    AU - Yatabe, Kohei

    AU - Oikawa, Yasuhiro

    PY - 2018/11/2

    Y1 - 2018/11/2

    N2 - In this paper, a sparsity-based method for solving an underde-termind source separation problem is proposed. The proposed method is formulated as a convex optimization problem with two kinds of sparsity priors: sparsity in time-frequency domain and direction-of-arrival (DOA). These priors enable simultaneous estimation of DOA and sound sources, while the estimation result does not depend on an initialization method thanks to the convexity. Experiments using 4 sound sources recorded by 2 microphones confirmed that every random initial value in the proposed method resulted in the same performance which was better than the conventional methods.

    AB - In this paper, a sparsity-based method for solving an underde-termind source separation problem is proposed. The proposed method is formulated as a convex optimization problem with two kinds of sparsity priors: sparsity in time-frequency domain and direction-of-arrival (DOA). These priors enable simultaneous estimation of DOA and sound sources, while the estimation result does not depend on an initialization method thanks to the convexity. Experiments using 4 sound sources recorded by 2 microphones confirmed that every random initial value in the proposed method resulted in the same performance which was better than the conventional methods.

    KW - 1-norm

    KW - 2-norm

    KW - Convex optimization

    KW - Disjointness

    KW - Group sparsity

    KW - L1

    KW - L2

    KW - Primal-dual splitting

    UR - http://www.scopus.com/inward/record.url?scp=85057363470&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85057363470&partnerID=8YFLogxK

    U2 - 10.1109/IWAENC.2018.8521359

    DO - 10.1109/IWAENC.2018.8521359

    M3 - Conference contribution

    AN - SCOPUS:85057363470

    SP - 161

    EP - 165

    BT - 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018 - Proceedings

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -