A combined order selection and parameter estimation algorithm for undamped exponentials

Ching Hui J. Ying; Ashutosh Sabharwal; Randolph L. Moses

doi:10.1109/78.824665

A combined order selection and parameter estimation algorithm for undamped exponentials

Ching Hui J. Ying, Ashutosh Sabharwal, Randolph L. Moses

Research output: Contribution to journal › Article › peer-review

27 Scopus citations

Abstract

We propose an approximate maximum likelihood parameter estimation algorithm, combined with a model order estimator, for superimposed undamped exponentials in noise. The algorithm combines the robustness of Fourier-based estimators and the high-resolution capabilities of parametric methods. We use a combination of a Wald statistic and a MAP test for order selection and initialize an iterative maximum likelihood descent algorithm recursively based on estimates at higher candidate model orders. Experiments using simulated data and synthetic radar data demonstrate improved performance over MDL, MAP, and AIC in cases of practical interest.

Original language	English (US)
Pages (from-to)	693-701
Number of pages	9
Journal	IEEE Transactions on Signal Processing
Volume	48
Issue number	3
DOIs	https://doi.org/10.1109/78.824665
State	Published - Mar 2000

Keywords

Combined detection and estimation
Resolution bounds, undamped exponentials

ASJC Scopus subject areas

Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/78.824665

Cite this

@article{bccb9897abdd482183ec649450f317de,

title = "A combined order selection and parameter estimation algorithm for undamped exponentials",

abstract = "We propose an approximate maximum likelihood parameter estimation algorithm, combined with a model order estimator, for superimposed undamped exponentials in noise. The algorithm combines the robustness of Fourier-based estimators and the high-resolution capabilities of parametric methods. We use a combination of a Wald statistic and a MAP test for order selection and initialize an iterative maximum likelihood descent algorithm recursively based on estimates at higher candidate model orders. Experiments using simulated data and synthetic radar data demonstrate improved performance over MDL, MAP, and AIC in cases of practical interest.",

keywords = "Combined detection and estimation, Resolution bounds, undamped exponentials",

author = "Ying, {Ching Hui J.} and Ashutosh Sabharwal and Moses, {Randolph L.}",

note = "Funding Information: Manuscript received March 19, 1998; revised February 15, 1999. This work was supported in part by the Defense Advanced Research Project Agency under Contract MDA972-93-1-0015 managed through Air Force Research Laboratories. The associate editor coordinating the review of this paper and approving it for publication was Prof. James A. Bucklew.",

year = "2000",

month = mar,

doi = "10.1109/78.824665",

language = "English (US)",

volume = "48",

pages = "693--701",

journal = "IEEE Transactions on Signal Processing",

issn = "1053-587X",

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

number = "3",

}

TY - JOUR

T1 - A combined order selection and parameter estimation algorithm for undamped exponentials

AU - Ying, Ching Hui J.

AU - Sabharwal, Ashutosh

AU - Moses, Randolph L.

N1 - Funding Information: Manuscript received March 19, 1998; revised February 15, 1999. This work was supported in part by the Defense Advanced Research Project Agency under Contract MDA972-93-1-0015 managed through Air Force Research Laboratories. The associate editor coordinating the review of this paper and approving it for publication was Prof. James A. Bucklew.

PY - 2000/3

Y1 - 2000/3

N2 - We propose an approximate maximum likelihood parameter estimation algorithm, combined with a model order estimator, for superimposed undamped exponentials in noise. The algorithm combines the robustness of Fourier-based estimators and the high-resolution capabilities of parametric methods. We use a combination of a Wald statistic and a MAP test for order selection and initialize an iterative maximum likelihood descent algorithm recursively based on estimates at higher candidate model orders. Experiments using simulated data and synthetic radar data demonstrate improved performance over MDL, MAP, and AIC in cases of practical interest.

AB - We propose an approximate maximum likelihood parameter estimation algorithm, combined with a model order estimator, for superimposed undamped exponentials in noise. The algorithm combines the robustness of Fourier-based estimators and the high-resolution capabilities of parametric methods. We use a combination of a Wald statistic and a MAP test for order selection and initialize an iterative maximum likelihood descent algorithm recursively based on estimates at higher candidate model orders. Experiments using simulated data and synthetic radar data demonstrate improved performance over MDL, MAP, and AIC in cases of practical interest.

KW - Combined detection and estimation

KW - Resolution bounds, undamped exponentials

UR - http://www.scopus.com/inward/record.url?scp=0033872153&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033872153&partnerID=8YFLogxK

U2 - 10.1109/78.824665

DO - 10.1109/78.824665

M3 - Article

AN - SCOPUS:0033872153

SN - 1053-587X

VL - 48

SP - 693

EP - 701

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

IS - 3

ER -

A combined order selection and parameter estimation algorithm for undamped exponentials

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this