Order statistics approach to estimation of the dimension of the noise subspace

Eran Fishler; Jonathan Friedman; Hagit Messer

doi:10.1109/HOST.1999.778743

Order statistics approach to estimation of the dimension of the noise subspace

Eran Fishler, Jonathan Friedman, Hagit Messer

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

Abstract

Model order selection and in particular determination of the dimension of the noise subspace, is an important problem in statistical signal processing. The discrete nature of the problem puts it in between detection and estimation. Standard tools from detection theory force a solution subject to arbitrary false alarm probability. On the other hand the direct maximum likelihood (ML) approach requires a penalty connection. In this paper we suggest the use of order statistics (OS) approach for the estimation of the dimension of the noise subspace. We show that the likelihood function of the ordered data has a unique non-trivial maximum with respect to the assumed dimension, and therefore we suggest an OS ML estimator. It is based on processing a single ordered sample and is, therefore, very simple. It assumes nothing about the distribution of the signal plus noise and therefore it is robust to the signal model. The suggested approach is demonstrated for i.i.d. exponential noise.

Original language	English
Title of host publication	Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	281-284
Number of pages	4
ISBN (Electronic)	0769501400, 9780769501406
DOIs	https://doi.org/10.1109/HOST.1999.778743
State	Published - 1999
Externally published	Yes
Event	1999 IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999 - Caesarea, Israel Duration: 14 Jun 1999 → 16 Jun 1999

Publication series

Name	Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999

Conference

Conference	1999 IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999
Country/Territory	Israel
City	Caesarea
Period	14/06/99 → 16/06/99

Bibliographical note

Publisher Copyright:
© 1999 IEEE.

Access to Document

10.1109/HOST.1999.778743

Cite this

Fishler, E., Friedman, J., & Messer, H. (1999). Order statistics approach to estimation of the dimension of the noise subspace. In Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999 (pp. 281-284). Article 778743 (Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/HOST.1999.778743

Fishler, Eran ; Friedman, Jonathan ; Messer, Hagit. / Order statistics approach to estimation of the dimension of the noise subspace. Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999. Institute of Electrical and Electronics Engineers Inc., 1999. pp. 281-284 (Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999).

@inproceedings{d2829ad2d455404d89567de6319b9aab,

title = "Order statistics approach to estimation of the dimension of the noise subspace",

abstract = "Model order selection and in particular determination of the dimension of the noise subspace, is an important problem in statistical signal processing. The discrete nature of the problem puts it in between detection and estimation. Standard tools from detection theory force a solution subject to arbitrary false alarm probability. On the other hand the direct maximum likelihood (ML) approach requires a penalty connection. In this paper we suggest the use of order statistics (OS) approach for the estimation of the dimension of the noise subspace. We show that the likelihood function of the ordered data has a unique non-trivial maximum with respect to the assumed dimension, and therefore we suggest an OS ML estimator. It is based on processing a single ordered sample and is, therefore, very simple. It assumes nothing about the distribution of the signal plus noise and therefore it is robust to the signal model. The suggested approach is demonstrated for i.i.d. exponential noise.",

author = "Eran Fishler and Jonathan Friedman and Hagit Messer",

note = "Publisher Copyright: {\textcopyright} 1999 IEEE.; 1999 IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999 ; Conference date: 14-06-1999 Through 16-06-1999",

year = "1999",

doi = "10.1109/HOST.1999.778743",

language = "אנגלית",

series = "Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "281--284",

booktitle = "Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999",

address = "ארצות הברית",

}

Fishler, E, Friedman, J & Messer, H 1999, Order statistics approach to estimation of the dimension of the noise subspace. in Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999., 778743, Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999, Institute of Electrical and Electronics Engineers Inc., pp. 281-284, 1999 IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999, Caesarea, Israel, 14/06/99. https://doi.org/10.1109/HOST.1999.778743

Order statistics approach to estimation of the dimension of the noise subspace. / Fishler, Eran; Friedman, Jonathan; Messer, Hagit.
Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999. Institute of Electrical and Electronics Engineers Inc., 1999. p. 281-284 778743 (Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999).

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

TY - GEN

T1 - Order statistics approach to estimation of the dimension of the noise subspace

AU - Fishler, Eran

AU - Friedman, Jonathan

AU - Messer, Hagit

PY - 1999

Y1 - 1999

N2 - Model order selection and in particular determination of the dimension of the noise subspace, is an important problem in statistical signal processing. The discrete nature of the problem puts it in between detection and estimation. Standard tools from detection theory force a solution subject to arbitrary false alarm probability. On the other hand the direct maximum likelihood (ML) approach requires a penalty connection. In this paper we suggest the use of order statistics (OS) approach for the estimation of the dimension of the noise subspace. We show that the likelihood function of the ordered data has a unique non-trivial maximum with respect to the assumed dimension, and therefore we suggest an OS ML estimator. It is based on processing a single ordered sample and is, therefore, very simple. It assumes nothing about the distribution of the signal plus noise and therefore it is robust to the signal model. The suggested approach is demonstrated for i.i.d. exponential noise.

AB - Model order selection and in particular determination of the dimension of the noise subspace, is an important problem in statistical signal processing. The discrete nature of the problem puts it in between detection and estimation. Standard tools from detection theory force a solution subject to arbitrary false alarm probability. On the other hand the direct maximum likelihood (ML) approach requires a penalty connection. In this paper we suggest the use of order statistics (OS) approach for the estimation of the dimension of the noise subspace. We show that the likelihood function of the ordered data has a unique non-trivial maximum with respect to the assumed dimension, and therefore we suggest an OS ML estimator. It is based on processing a single ordered sample and is, therefore, very simple. It assumes nothing about the distribution of the signal plus noise and therefore it is robust to the signal model. The suggested approach is demonstrated for i.i.d. exponential noise.

UR - http://www.scopus.com/inward/record.url?scp=17244364521&partnerID=8YFLogxK

U2 - 10.1109/HOST.1999.778743

DO - 10.1109/HOST.1999.778743

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:17244364521

T3 - Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999

SP - 281

EP - 284

BT - Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1999 IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999

Y2 - 14 June 1999 through 16 June 1999

ER -

Fishler E, Friedman J, Messer H. Order statistics approach to estimation of the dimension of the noise subspace. In Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999. Institute of Electrical and Electronics Engineers Inc. 1999. p. 281-284. 778743. (Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics, SPW-HOS 1999). doi: 10.1109/HOST.1999.778743

Order statistics approach to estimation of the dimension of the noise subspace

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this