Recursive 4SID algorithms using gradient type subspace tracking

H. Oku, Hidenori Kimura

    Research output: Contribution to journalArticleAcademic

    72 Citations (Scopus)

    Abstract

    Sometimes we obtain some prior information about a system to be identified, e.g., the order, model structure etc. In this paper, we consider the case where the order of a MIMO system to be identified is a priori known. Recursive subspace state-space system identification algorithms presented here are based on the gradient type subspace tracking method used in the array signal processing. The algorithms enable us to estimate directly the subspace spanned by the column vectors of the extended observability matrix of the system to be identified without performing the singular value decomposition. Also, a new convergence proof of the gradient type subspace tracking is given in this paper. Under the condition of a step size between 0 and 1, we prove the convergence property of the recursive equation of the gradient type subspace tracking. A numerical example illustrates that our algorithm is more robust with respect to the choice of the initial values than the corresponding PAST one.
    Original languageUndefined
    Pages (from-to)1035-1043
    JournalAutomatica
    Volume38
    Issue number6
    DOIs
    Publication statusPublished - 2002

    Keywords

    • Matrix inversion
    • Convergence proofs
    • Subspace methods
    • Gradient methods
    • IR-74736
    • Adaptive filters
    • Identification algorithms

    Cite this

    Oku, H. ; Kimura, Hidenori. / Recursive 4SID algorithms using gradient type subspace tracking. In: Automatica. 2002 ; Vol. 38, No. 6. pp. 1035-1043.
    @article{5cac85286295482da34d274ee2526c7c,
    title = "Recursive 4SID algorithms using gradient type subspace tracking",
    abstract = "Sometimes we obtain some prior information about a system to be identified, e.g., the order, model structure etc. In this paper, we consider the case where the order of a MIMO system to be identified is a priori known. Recursive subspace state-space system identification algorithms presented here are based on the gradient type subspace tracking method used in the array signal processing. The algorithms enable us to estimate directly the subspace spanned by the column vectors of the extended observability matrix of the system to be identified without performing the singular value decomposition. Also, a new convergence proof of the gradient type subspace tracking is given in this paper. Under the condition of a step size between 0 and 1, we prove the convergence property of the recursive equation of the gradient type subspace tracking. A numerical example illustrates that our algorithm is more robust with respect to the choice of the initial values than the corresponding PAST one.",
    keywords = "Matrix inversion, Convergence proofs, Subspace methods, Gradient methods, IR-74736, Adaptive filters, Identification algorithms",
    author = "H. Oku and Hidenori Kimura",
    year = "2002",
    doi = "10.1016/S0005-1098(01)00286-2",
    language = "Undefined",
    volume = "38",
    pages = "1035--1043",
    journal = "Automatica",
    issn = "0005-1098",
    publisher = "Elsevier",
    number = "6",

    }

    Recursive 4SID algorithms using gradient type subspace tracking. / Oku, H.; Kimura, Hidenori.

    In: Automatica, Vol. 38, No. 6, 2002, p. 1035-1043.

    Research output: Contribution to journalArticleAcademic

    TY - JOUR

    T1 - Recursive 4SID algorithms using gradient type subspace tracking

    AU - Oku, H.

    AU - Kimura, Hidenori

    PY - 2002

    Y1 - 2002

    N2 - Sometimes we obtain some prior information about a system to be identified, e.g., the order, model structure etc. In this paper, we consider the case where the order of a MIMO system to be identified is a priori known. Recursive subspace state-space system identification algorithms presented here are based on the gradient type subspace tracking method used in the array signal processing. The algorithms enable us to estimate directly the subspace spanned by the column vectors of the extended observability matrix of the system to be identified without performing the singular value decomposition. Also, a new convergence proof of the gradient type subspace tracking is given in this paper. Under the condition of a step size between 0 and 1, we prove the convergence property of the recursive equation of the gradient type subspace tracking. A numerical example illustrates that our algorithm is more robust with respect to the choice of the initial values than the corresponding PAST one.

    AB - Sometimes we obtain some prior information about a system to be identified, e.g., the order, model structure etc. In this paper, we consider the case where the order of a MIMO system to be identified is a priori known. Recursive subspace state-space system identification algorithms presented here are based on the gradient type subspace tracking method used in the array signal processing. The algorithms enable us to estimate directly the subspace spanned by the column vectors of the extended observability matrix of the system to be identified without performing the singular value decomposition. Also, a new convergence proof of the gradient type subspace tracking is given in this paper. Under the condition of a step size between 0 and 1, we prove the convergence property of the recursive equation of the gradient type subspace tracking. A numerical example illustrates that our algorithm is more robust with respect to the choice of the initial values than the corresponding PAST one.

    KW - Matrix inversion

    KW - Convergence proofs

    KW - Subspace methods

    KW - Gradient methods

    KW - IR-74736

    KW - Adaptive filters

    KW - Identification algorithms

    U2 - 10.1016/S0005-1098(01)00286-2

    DO - 10.1016/S0005-1098(01)00286-2

    M3 - Article

    VL - 38

    SP - 1035

    EP - 1043

    JO - Automatica

    JF - Automatica

    SN - 0005-1098

    IS - 6

    ER -