Advanced

Computationally efficient damped Capon and APES spectral estimation

Glentis, George-Othan and Jakobsson, Andreas LU (2013) 21st European Signal Processing Conference (EUSIPCO 2013)
Abstract
In this paper, we introduce computationally efficient imple-

mentations of the data-dependent non-parametric damped

Capon (dCapon) and APES (dAPES) spectral estimators.

These estimators form two-dimensional frequency represen-

tations over both frequency and damping, and have been

shown to enable an efficient separation of closely spaced

spectral lines with different line widths. The proposed imple-

mentations are formed using an FFT-based fast polynomial

reformulation exploiting the displacement structure of the

matrices associated with the trigonometric polynomials that

appear in the nominator and the denominator of the spec-

tral... (More)
In this paper, we introduce computationally efficient imple-

mentations of the data-dependent non-parametric damped

Capon (dCapon) and APES (dAPES) spectral estimators.

These estimators form two-dimensional frequency represen-

tations over both frequency and damping, and have been

shown to enable an efficient separation of closely spaced

spectral lines with different line widths. The proposed imple-

mentations are formed using an FFT-based fast polynomial

reformulation exploiting the displacement structure of the

matrices associated with the trigonometric polynomials that

appear in the nominator and the denominator of the spec-

tral estimators. The resulting implementations are exact and

notably reduces the required computational complexity of

the estimators. Numerical simulations illustrates the perfor-

mance and the achieved complexity gain of the proposed

implementations (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
in press
subject
keywords
Spectral estimation, damped sinusoidals, Capon, APES, computationally efficient implementations
conference name
21st European Signal Processing Conference (EUSIPCO 2013)
language
English
LU publication?
yes
id
99be0a1c-58d9-46fb-a342-e9ba569b642f (old id 3991197)
date added to LUP
2014-01-17 19:09:57
date last changed
2016-07-08 13:47:51
@misc{99be0a1c-58d9-46fb-a342-e9ba569b642f,
  abstract     = {In this paper, we introduce computationally efficient imple-<br/><br>
mentations of the data-dependent non-parametric damped<br/><br>
Capon (dCapon) and APES (dAPES) spectral estimators.<br/><br>
These estimators form two-dimensional frequency represen-<br/><br>
tations over both frequency and damping, and have been<br/><br>
shown to enable an efficient separation of closely spaced<br/><br>
spectral lines with different line widths. The proposed imple-<br/><br>
mentations are formed using an FFT-based fast polynomial<br/><br>
reformulation exploiting the displacement structure of the<br/><br>
matrices associated with the trigonometric polynomials that<br/><br>
appear in the nominator and the denominator of the spec-<br/><br>
tral estimators. The resulting implementations are exact and<br/><br>
notably reduces the required computational complexity of<br/><br>
the estimators. Numerical simulations illustrates the perfor-<br/><br>
mance and the achieved complexity gain of the proposed<br/><br>
implementations},
  author       = {Glentis, George-Othan and Jakobsson, Andreas},
  keyword      = {Spectral estimation,damped sinusoidals,Capon,APES,computationally efficient implementations},
  language     = {eng},
  title        = {Computationally efficient damped Capon and APES spectral estimation},
  year         = {2013},
}