ℓ1-regularized recursive total least squares based sparse system identification for the error-in-variables

In this paper an ℓ 1 -regularized recursive total least squares (RTLS) algorithm is considered for the sparse system identification. Although recursive least squares (RLS) has been successfully applied in sparse system identification, the estimation performance in RLS based algorithms becomes worse,...

Full description

Saved in:
Bibliographic Details
Published in:SpringerPlus Vol. 5; no. 1
Main Authors: Lim, Jun-seok, Pang, Hee-Suk
Format: Journal Article
Language:English
Published: Cham Springer International Publishing 31.08.2016
Subjects:
Engineering
Humanities and Social Sciences
multidisciplinary
Science
Science (multidisciplinary)
RLS
Convex regularization
TLS
1-norm
Adaptive filter
Sparsity
ISSN:2193-1801
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper an ℓ 1 -regularized recursive total least squares (RTLS) algorithm is considered for the sparse system identification. Although recursive least squares (RLS) has been successfully applied in sparse system identification, the estimation performance in RLS based algorithms becomes worse, when both input and output are contaminated by noise (the error-in-variables problem). We proposed an algorithm to handle the error-in-variables problem. The proposed ℓ 1 -RTLS algorithm is an RLS like iteration using the ℓ 1 regularization. The proposed algorithm not only gives excellent performance but also reduces the required complexity through the effective inversion matrix handling. Simulations demonstrate the superiority of the proposed ℓ 1 -regularized RTLS for the sparse system identification setting.
ISSN:2193-1801
DOI:10.1186/s40064-016-3120-6