Compressive sampling for signal classification

J. Haupt; R.M. Castro; R. Nowak; G. Fudge; A. Yeh

doi:10.1109/ACSSC.2006.354994

Compressive sampling for signal classification

J. Haupt, R.M. Castro, R. Nowak, G. Fudge, A. Yeh

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

102 Citations (Scopus)

1 Downloads (Pure)

Abstract

Compressive sampling (CS), also called compressed sensing, entails making observations of an unknown signal by projecting it onto random vectors. Recent theoretical results show that if the signal is sparse (or nearly sparse) in some basis, then with high probability such observations essentially encode the salient information in the signal. Further, the signal can be reconstructed from these "random projections," even when the number of observations is far less than the ambient signal dimension. The provable success of CS for signal reconstruction motivates the study of its potential in other applications. This paper investigates the utility of CS projection observations for signal classification (more specifically, m-ary hypothesis testing). Theoretical error bounds are derived and verified with several simulations.

Original language	English
Title of host publication	Proceedings 40th Asilomar Conference on Signals, Systems and Computers (ACSSC'06, Pacific Grove CA, USA, October 29-November 6, 2006)
Publisher	Institute of Electrical and Electronics Engineers
Pages	1430-1434
ISBN (Print)	1-4244-0784-2
DOIs	https://doi.org/10.1109/ACSSC.2006.354994
Publication status	Published - 2006

Access to Document

10.1109/ACSSC.2006.354994

Cite this

@inproceedings{cc47a6b80dd24f9c8a5508f0f2031b68,

title = "Compressive sampling for signal classification",

abstract = "Compressive sampling (CS), also called compressed sensing, entails making observations of an unknown signal by projecting it onto random vectors. Recent theoretical results show that if the signal is sparse (or nearly sparse) in some basis, then with high probability such observations essentially encode the salient information in the signal. Further, the signal can be reconstructed from these {"}random projections,{"} even when the number of observations is far less than the ambient signal dimension. The provable success of CS for signal reconstruction motivates the study of its potential in other applications. This paper investigates the utility of CS projection observations for signal classification (more specifically, m-ary hypothesis testing). Theoretical error bounds are derived and verified with several simulations.",

author = "J. Haupt and R.M. Castro and R. Nowak and G. Fudge and A. Yeh",

year = "2006",

doi = "10.1109/ACSSC.2006.354994",

language = "English",

isbn = "1-4244-0784-2",

pages = "1430--1434",

booktitle = "Proceedings 40th Asilomar Conference on Signals, Systems and Computers (ACSSC'06, Pacific Grove CA, USA, October 29-November 6, 2006)",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

}

Compressive sampling for signal classification. / Haupt, J.; Castro, R.M.; Nowak, R. et al.
Proceedings 40th Asilomar Conference on Signals, Systems and Computers (ACSSC'06, Pacific Grove CA, USA, October 29-November 6, 2006). Institute of Electrical and Electronics Engineers, 2006. p. 1430-1434.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Compressive sampling for signal classification

AU - Haupt, J.

AU - Castro, R.M.

AU - Nowak, R.

AU - Fudge, G.

AU - Yeh, A.

PY - 2006

Y1 - 2006

N2 - Compressive sampling (CS), also called compressed sensing, entails making observations of an unknown signal by projecting it onto random vectors. Recent theoretical results show that if the signal is sparse (or nearly sparse) in some basis, then with high probability such observations essentially encode the salient information in the signal. Further, the signal can be reconstructed from these "random projections," even when the number of observations is far less than the ambient signal dimension. The provable success of CS for signal reconstruction motivates the study of its potential in other applications. This paper investigates the utility of CS projection observations for signal classification (more specifically, m-ary hypothesis testing). Theoretical error bounds are derived and verified with several simulations.

AB - Compressive sampling (CS), also called compressed sensing, entails making observations of an unknown signal by projecting it onto random vectors. Recent theoretical results show that if the signal is sparse (or nearly sparse) in some basis, then with high probability such observations essentially encode the salient information in the signal. Further, the signal can be reconstructed from these "random projections," even when the number of observations is far less than the ambient signal dimension. The provable success of CS for signal reconstruction motivates the study of its potential in other applications. This paper investigates the utility of CS projection observations for signal classification (more specifically, m-ary hypothesis testing). Theoretical error bounds are derived and verified with several simulations.

U2 - 10.1109/ACSSC.2006.354994

DO - 10.1109/ACSSC.2006.354994

M3 - Conference contribution

SN - 1-4244-0784-2

SP - 1430

EP - 1434

BT - Proceedings 40th Asilomar Conference on Signals, Systems and Computers (ACSSC'06, Pacific Grove CA, USA, October 29-November 6, 2006)

PB - Institute of Electrical and Electronics Engineers

ER -

Compressive sampling for signal classification

Abstract

Access to Document

Fingerprint

Cite this