Online Sparse Volterra System Identification Using Projections onto Weighted <I>l</I><SUB>1</SUB> Balls

Tae-Ho JUNG; Jung-Hee KIM; Joon-Hyuk CHANG; Sang Won NAM

doi:10.1587/transfun.E96.A.1980

Online Sparse Volterra System Identification Using Projections onto Weighted l₁ Balls

Tae-Ho JUNG, Jung-Hee KIM, Joon-Hyuk CHANG, Sang Won NAM

Full Text Views

0

Share
Cite this

Summary :

In this paper, online sparse Volterra system identification is proposed. For that purpose, the conventional adaptive projection-based algorithm with weighted l₁ balls (APWL1) is revisited for nonlinear system identification, whereby the linear-in-parameters nature of Volterra systems is utilized. Compared with sparsity-aware recursive least squares (RLS) based algorithms, requiring higher computational complexity and showing faster convergence and lower steady-state error due to their long memory in time-invariant cases, the proposed approach yields better tracking capability in time-varying cases due to short-term data dependence in updating the weight. Also, when N is the number of sparse Volterra kernels and q is the number of input vectors involved to update the weight, the proposed algorithm requires O(qN) multiplication complexity and O(Nlog ₂N) sorting-operation complexity. Furthermore, sparsity-aware least mean-squares and affine projection based algorithms are also tested.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E96-A No.10 pp.1980-1983

Publication Date: 2013/10/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E96.A.1980

Type of Manuscript: Special Section PAPER (Special Section on Sparsity-aware Signal Processing)

Category

Authors

Tae-Ho JUNG
  Hanyang University
Jung-Hee KIM
  Hanyang University
Joon-Hyuk CHANG
  Hanyang University
Sang Won NAM
  Hanyang University

Keyword

adaptive filtering, sparse Volterra systems, identification, projections

Cite this

Copy

Tae-Ho JUNG, Jung-Hee KIM, Joon-Hyuk CHANG, Sang Won NAM, "Online Sparse Volterra System Identification Using Projections onto Weighted l1 Balls" in IEICE TRANSACTIONS on Fundamentals, vol. E96-A, no. 10, pp. 1980-1983, October 2013, doi: 10.1587/transfun.E96.A.1980.
Abstract: In this paper, online sparse Volterra system identification is proposed. For that purpose, the conventional adaptive projection-based algorithm with weighted l₁ balls (APWL1) is revisited for nonlinear system identification, whereby the linear-in-parameters nature of Volterra systems is utilized. Compared with sparsity-aware recursive least squares (RLS) based algorithms, requiring higher computational complexity and showing faster convergence and lower steady-state error due to their long memory in time-invariant cases, the proposed approach yields better tracking capability in time-varying cases due to short-term data dependence in updating the weight. Also, when N is the number of sparse Volterra kernels and q is the number of input vectors involved to update the weight, the proposed algorithm requires O(qN) multiplication complexity and O(Nlog ₂N) sorting-operation complexity. Furthermore, sparsity-aware least mean-squares and affine projection based algorithms are also tested.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/transfun.E96.A.1980/_p

Copy

@ARTICLE{e96-a_10_1980,
author={Tae-Ho JUNG, Jung-Hee KIM, Joon-Hyuk CHANG, Sang Won NAM, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Online Sparse Volterra System Identification Using Projections onto Weighted l1 Balls},
year={2013},
volume={E96-A},
number={10},
pages={1980-1983},
abstract={In this paper, online sparse Volterra system identification is proposed. For that purpose, the conventional adaptive projection-based algorithm with weighted l₁ balls (APWL1) is revisited for nonlinear system identification, whereby the linear-in-parameters nature of Volterra systems is utilized. Compared with sparsity-aware recursive least squares (RLS) based algorithms, requiring higher computational complexity and showing faster convergence and lower steady-state error due to their long memory in time-invariant cases, the proposed approach yields better tracking capability in time-varying cases due to short-term data dependence in updating the weight. Also, when N is the number of sparse Volterra kernels and q is the number of input vectors involved to update the weight, the proposed algorithm requires O(qN) multiplication complexity and O(Nlog ₂N) sorting-operation complexity. Furthermore, sparsity-aware least mean-squares and affine projection based algorithms are also tested.},
keywords={},
doi={10.1587/transfun.E96.A.1980},
ISSN={1745-1337},
month={October},}

Copy

TY - JOUR
TI - Online Sparse Volterra System Identification Using Projections onto Weighted l1 Balls
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1980
EP - 1983
AU - Tae-Ho JUNG
AU - Jung-Hee KIM
AU - Joon-Hyuk CHANG
AU - Sang Won NAM
PY - 2013
DO - 10.1587/transfun.E96.A.1980
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E96-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 2013
AB - In this paper, online sparse Volterra system identification is proposed. For that purpose, the conventional adaptive projection-based algorithm with weighted l₁ balls (APWL1) is revisited for nonlinear system identification, whereby the linear-in-parameters nature of Volterra systems is utilized. Compared with sparsity-aware recursive least squares (RLS) based algorithms, requiring higher computational complexity and showing faster convergence and lower steady-state error due to their long memory in time-invariant cases, the proposed approach yields better tracking capability in time-varying cases due to short-term data dependence in updating the weight. Also, when N is the number of sparse Volterra kernels and q is the number of input vectors involved to update the weight, the proposed algorithm requires O(qN) multiplication complexity and O(Nlog ₂N) sorting-operation complexity. Furthermore, sparsity-aware least mean-squares and affine projection based algorithms are also tested.
ER -