Compressed Sensing with Combinatorial Designs

Theory and Simulations

Darryn Bryant, Charles Colbourn, Daniel Horsley, Padraig O Cathain

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We use deterministic and probabilistic methods to analyze the performance of compressed sensing matrices constructed from Hadamard matrices and pairwise balanced designs, previously introduced by a subset of the authors. In this paper, we obtain upper and lower bounds on the sparsity of signals for which our matrices guarantee recovery. These bounds are tight to within a multiplicative factor of at most. We provide new theoretical results and detailed simulations, which indicate that the construction is competitive with Gaussian random matrices, and that recovery is tolerant to noise. A new recovery algorithm tailored to the construction is also given.

Original languageEnglish (US)
Article number7954982
Pages (from-to)4850-4859
Number of pages10
JournalIEEE Transactions on Information Theory
Volume63
Issue number8
DOIs
StatePublished - Aug 1 2017

Fingerprint

Compressed sensing
Recovery
simulation
Hadamard matrices
guarantee
performance

Keywords

  • Combinatorial designs
  • Compressed sensing
  • Signal recovery

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences

Cite this

Compressed Sensing with Combinatorial Designs : Theory and Simulations. / Bryant, Darryn; Colbourn, Charles; Horsley, Daniel; O Cathain, Padraig.

In: IEEE Transactions on Information Theory, Vol. 63, No. 8, 7954982, 01.08.2017, p. 4850-4859.

Research output: Contribution to journalArticle

Bryant, Darryn ; Colbourn, Charles ; Horsley, Daniel ; O Cathain, Padraig. / Compressed Sensing with Combinatorial Designs : Theory and Simulations. In: IEEE Transactions on Information Theory. 2017 ; Vol. 63, No. 8. pp. 4850-4859.
@article{1e50965838bd41e88472ae969d3c72f8,
title = "Compressed Sensing with Combinatorial Designs: Theory and Simulations",
abstract = "We use deterministic and probabilistic methods to analyze the performance of compressed sensing matrices constructed from Hadamard matrices and pairwise balanced designs, previously introduced by a subset of the authors. In this paper, we obtain upper and lower bounds on the sparsity of signals for which our matrices guarantee recovery. These bounds are tight to within a multiplicative factor of at most. We provide new theoretical results and detailed simulations, which indicate that the construction is competitive with Gaussian random matrices, and that recovery is tolerant to noise. A new recovery algorithm tailored to the construction is also given.",
keywords = "Combinatorial designs, Compressed sensing, Signal recovery",
author = "Darryn Bryant and Charles Colbourn and Daniel Horsley and {O Cathain}, Padraig",
year = "2017",
month = "8",
day = "1",
doi = "10.1109/TIT.2017.2717584",
language = "English (US)",
volume = "63",
pages = "4850--4859",
journal = "IEEE Transactions on Information Theory",
issn = "0018-9448",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - Compressed Sensing with Combinatorial Designs

T2 - Theory and Simulations

AU - Bryant, Darryn

AU - Colbourn, Charles

AU - Horsley, Daniel

AU - O Cathain, Padraig

PY - 2017/8/1

Y1 - 2017/8/1

N2 - We use deterministic and probabilistic methods to analyze the performance of compressed sensing matrices constructed from Hadamard matrices and pairwise balanced designs, previously introduced by a subset of the authors. In this paper, we obtain upper and lower bounds on the sparsity of signals for which our matrices guarantee recovery. These bounds are tight to within a multiplicative factor of at most. We provide new theoretical results and detailed simulations, which indicate that the construction is competitive with Gaussian random matrices, and that recovery is tolerant to noise. A new recovery algorithm tailored to the construction is also given.

AB - We use deterministic and probabilistic methods to analyze the performance of compressed sensing matrices constructed from Hadamard matrices and pairwise balanced designs, previously introduced by a subset of the authors. In this paper, we obtain upper and lower bounds on the sparsity of signals for which our matrices guarantee recovery. These bounds are tight to within a multiplicative factor of at most. We provide new theoretical results and detailed simulations, which indicate that the construction is competitive with Gaussian random matrices, and that recovery is tolerant to noise. A new recovery algorithm tailored to the construction is also given.

KW - Combinatorial designs

KW - Compressed sensing

KW - Signal recovery

UR - http://www.scopus.com/inward/record.url?scp=85023766965&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023766965&partnerID=8YFLogxK

U2 - 10.1109/TIT.2017.2717584

DO - 10.1109/TIT.2017.2717584

M3 - Article

VL - 63

SP - 4850

EP - 4859

JO - IEEE Transactions on Information Theory

JF - IEEE Transactions on Information Theory

SN - 0018-9448

IS - 8

M1 - 7954982

ER -