Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm

Riyanto T. Bambang, Redi R. Yacoub, Kenko Uchida

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

    11 Citations (Scopus)

    Abstract

    This paper presents theoretical and experimental modeling of secondary path of an active noise control system in free space by using recurrent neural networks. A learning algorithm for diagonal recurrent neural networks is developed based on Extended Kalman Filter and is referred to as Diagonal Recurrent Extended Kalman Filter algorithm. The neural network structure and its algorithm are applied to handle nonlinearity of the secondary path. To put the neural identification task within the context of ANC, a new control algorithm based on DREKF is also presented. The real-time experiment, however, is performed only for identification task. Experimental results using a floating point DSP show that the number of neurons in neural network can be reduced by introducing the diagonal recurrent elements, without deteriorating the identification system performance.

    Original languageEnglish
    Title of host publication2004 5th Asian Control Conference
    Pages665-673
    Number of pages9
    Volume1
    Publication statusPublished - 2004
    Event2004 5th Asian Control Conference - Melbourne
    Duration: 2004 Jul 202004 Jul 23

    Other

    Other2004 5th Asian Control Conference
    CityMelbourne
    Period04/7/2004/7/23

    Fingerprint

    Recurrent neural networks
    Extended Kalman filters
    Neural networks
    Active noise control
    Learning algorithms
    Neurons
    Identification (control systems)
    Control systems
    Experiments

    Keywords

    • ANC
    • Diagonal recurrent neural networks
    • DSP
    • Extended Kalman Filter
    • Identification
    • Secondary path nonlinearity

    ASJC Scopus subject areas

    • Engineering(all)

    Cite this

    Bambang, R. T., Yacoub, R. R., & Uchida, K. (2004). Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm. In 2004 5th Asian Control Conference (Vol. 1, pp. 665-673)

    Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm. / Bambang, Riyanto T.; Yacoub, Redi R.; Uchida, Kenko.

    2004 5th Asian Control Conference. Vol. 1 2004. p. 665-673.

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

    Bambang, RT, Yacoub, RR & Uchida, K 2004, Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm. in 2004 5th Asian Control Conference. vol. 1, pp. 665-673, 2004 5th Asian Control Conference, Melbourne, 04/7/20.
    Bambang RT, Yacoub RR, Uchida K. Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm. In 2004 5th Asian Control Conference. Vol. 1. 2004. p. 665-673
    Bambang, Riyanto T. ; Yacoub, Redi R. ; Uchida, Kenko. / Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm. 2004 5th Asian Control Conference. Vol. 1 2004. pp. 665-673
    @inproceedings{fc623974dc8445238547178fa1882a40,
    title = "Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm",
    abstract = "This paper presents theoretical and experimental modeling of secondary path of an active noise control system in free space by using recurrent neural networks. A learning algorithm for diagonal recurrent neural networks is developed based on Extended Kalman Filter and is referred to as Diagonal Recurrent Extended Kalman Filter algorithm. The neural network structure and its algorithm are applied to handle nonlinearity of the secondary path. To put the neural identification task within the context of ANC, a new control algorithm based on DREKF is also presented. The real-time experiment, however, is performed only for identification task. Experimental results using a floating point DSP show that the number of neurons in neural network can be reduced by introducing the diagonal recurrent elements, without deteriorating the identification system performance.",
    keywords = "ANC, Diagonal recurrent neural networks, DSP, Extended Kalman Filter, Identification, Secondary path nonlinearity",
    author = "Bambang, {Riyanto T.} and Yacoub, {Redi R.} and Kenko Uchida",
    year = "2004",
    language = "English",
    isbn = "0780388739",
    volume = "1",
    pages = "665--673",
    booktitle = "2004 5th Asian Control Conference",

    }

    TY - GEN

    T1 - Identification of secondary path in ANC using diagonal recurrent neural networks with EKF algorithm

    AU - Bambang, Riyanto T.

    AU - Yacoub, Redi R.

    AU - Uchida, Kenko

    PY - 2004

    Y1 - 2004

    N2 - This paper presents theoretical and experimental modeling of secondary path of an active noise control system in free space by using recurrent neural networks. A learning algorithm for diagonal recurrent neural networks is developed based on Extended Kalman Filter and is referred to as Diagonal Recurrent Extended Kalman Filter algorithm. The neural network structure and its algorithm are applied to handle nonlinearity of the secondary path. To put the neural identification task within the context of ANC, a new control algorithm based on DREKF is also presented. The real-time experiment, however, is performed only for identification task. Experimental results using a floating point DSP show that the number of neurons in neural network can be reduced by introducing the diagonal recurrent elements, without deteriorating the identification system performance.

    AB - This paper presents theoretical and experimental modeling of secondary path of an active noise control system in free space by using recurrent neural networks. A learning algorithm for diagonal recurrent neural networks is developed based on Extended Kalman Filter and is referred to as Diagonal Recurrent Extended Kalman Filter algorithm. The neural network structure and its algorithm are applied to handle nonlinearity of the secondary path. To put the neural identification task within the context of ANC, a new control algorithm based on DREKF is also presented. The real-time experiment, however, is performed only for identification task. Experimental results using a floating point DSP show that the number of neurons in neural network can be reduced by introducing the diagonal recurrent elements, without deteriorating the identification system performance.

    KW - ANC

    KW - Diagonal recurrent neural networks

    KW - DSP

    KW - Extended Kalman Filter

    KW - Identification

    KW - Secondary path nonlinearity

    UR - http://www.scopus.com/inward/record.url?scp=16244367462&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=16244367462&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:16244367462

    SN - 0780388739

    VL - 1

    SP - 665

    EP - 673

    BT - 2004 5th Asian Control Conference

    ER -