Projection-based and look-ahead strategies for atom selection

Saikat Chatterjee, Dennis Sundman, Mikko Vehkaperä, Mikael Skoglund

Research output: Contribution to journalArticleScientificpeer-review

37 Citations (Scopus)

Abstract

In this paper, we improve iterative greedy search algorithms in which atoms are selected serially over iterations, i.e., one-by-one over iterations. For serial atom selection, we devise two new schemes to select an atom from a set of potential atoms in each iteration. The two new schemes lead to two new algorithms. For both the algorithms, in each iteration, the set of potential atoms is found using a standard matched filter. In case of the first scheme, we propose an orthogonal projection strategy that selects an atom from the set of potential atoms. Then, for the second scheme, we propose a look-ahead strategy such that the selection of an atom in the current iteration has an effect on the future iterations. The use of look-ahead strategy requires a higher computational resource. To achieve a tradeoff between performance and complexity, we use the two new schemes in cascade and develop a third new algorithm. Through experimental evaluations, we compare the proposed algorithms with existing greedy search and convex relaxation algorithms.
Original languageEnglish
Article number6062423
Pages (from-to)634-647
JournalIEEE Transactions on Signal Processing
Volume60
Issue number2
DOIs
Publication statusPublished - 1 Feb 2012
MoE publication typeA1 Journal article-refereed

Fingerprint

Atoms
Matched filters

Keywords

  • Compressive sampling
  • orthogonal least squares
  • orthogonal matching pursuit
  • sparse signal estimation
  • subspace pursuit

Cite this

Chatterjee, Saikat ; Sundman, Dennis ; Vehkaperä, Mikko ; Skoglund, Mikael. / Projection-based and look-ahead strategies for atom selection. In: IEEE Transactions on Signal Processing. 2012 ; Vol. 60, No. 2. pp. 634-647.
@article{0b038029bb2b4322b465df2ac5b5d132,
title = "Projection-based and look-ahead strategies for atom selection",
abstract = "In this paper, we improve iterative greedy search algorithms in which atoms are selected serially over iterations, i.e., one-by-one over iterations. For serial atom selection, we devise two new schemes to select an atom from a set of potential atoms in each iteration. The two new schemes lead to two new algorithms. For both the algorithms, in each iteration, the set of potential atoms is found using a standard matched filter. In case of the first scheme, we propose an orthogonal projection strategy that selects an atom from the set of potential atoms. Then, for the second scheme, we propose a look-ahead strategy such that the selection of an atom in the current iteration has an effect on the future iterations. The use of look-ahead strategy requires a higher computational resource. To achieve a tradeoff between performance and complexity, we use the two new schemes in cascade and develop a third new algorithm. Through experimental evaluations, we compare the proposed algorithms with existing greedy search and convex relaxation algorithms.",
keywords = "Compressive sampling, orthogonal least squares, orthogonal matching pursuit, sparse signal estimation, subspace pursuit",
author = "Saikat Chatterjee and Dennis Sundman and Mikko Vehkaper{\"a} and Mikael Skoglund",
year = "2012",
month = "2",
day = "1",
doi = "10.1109/TSP.2011.2173682",
language = "English",
volume = "60",
pages = "634--647",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "IEEE Institute of Electrical and Electronic Engineers",
number = "2",

}

Projection-based and look-ahead strategies for atom selection. / Chatterjee, Saikat; Sundman, Dennis; Vehkaperä, Mikko; Skoglund, Mikael.

In: IEEE Transactions on Signal Processing, Vol. 60, No. 2, 6062423, 01.02.2012, p. 634-647.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Projection-based and look-ahead strategies for atom selection

AU - Chatterjee, Saikat

AU - Sundman, Dennis

AU - Vehkaperä, Mikko

AU - Skoglund, Mikael

PY - 2012/2/1

Y1 - 2012/2/1

N2 - In this paper, we improve iterative greedy search algorithms in which atoms are selected serially over iterations, i.e., one-by-one over iterations. For serial atom selection, we devise two new schemes to select an atom from a set of potential atoms in each iteration. The two new schemes lead to two new algorithms. For both the algorithms, in each iteration, the set of potential atoms is found using a standard matched filter. In case of the first scheme, we propose an orthogonal projection strategy that selects an atom from the set of potential atoms. Then, for the second scheme, we propose a look-ahead strategy such that the selection of an atom in the current iteration has an effect on the future iterations. The use of look-ahead strategy requires a higher computational resource. To achieve a tradeoff between performance and complexity, we use the two new schemes in cascade and develop a third new algorithm. Through experimental evaluations, we compare the proposed algorithms with existing greedy search and convex relaxation algorithms.

AB - In this paper, we improve iterative greedy search algorithms in which atoms are selected serially over iterations, i.e., one-by-one over iterations. For serial atom selection, we devise two new schemes to select an atom from a set of potential atoms in each iteration. The two new schemes lead to two new algorithms. For both the algorithms, in each iteration, the set of potential atoms is found using a standard matched filter. In case of the first scheme, we propose an orthogonal projection strategy that selects an atom from the set of potential atoms. Then, for the second scheme, we propose a look-ahead strategy such that the selection of an atom in the current iteration has an effect on the future iterations. The use of look-ahead strategy requires a higher computational resource. To achieve a tradeoff between performance and complexity, we use the two new schemes in cascade and develop a third new algorithm. Through experimental evaluations, we compare the proposed algorithms with existing greedy search and convex relaxation algorithms.

KW - Compressive sampling

KW - orthogonal least squares

KW - orthogonal matching pursuit

KW - sparse signal estimation

KW - subspace pursuit

UR - http://www.scopus.com/inward/record.url?scp=84855955910&partnerID=8YFLogxK

U2 - 10.1109/TSP.2011.2173682

DO - 10.1109/TSP.2011.2173682

M3 - Article

AN - SCOPUS:84855955910

VL - 60

SP - 634

EP - 647

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 2

M1 - 6062423

ER -