2015
DOI: 10.1109/tsp.2015.2420541
|View full text |Cite
|
Sign up to set email alerts
|

On Gridless Sparse Methods for Line Spectral Estimation From Complete and Incomplete Data

Abstract: Abstract-This paper is concerned about sparse, continuous frequency estimation in line spectral estimation, and focused on developing gridless sparse methods which overcome grid mismatches and correspond to limiting scenarios of existing grid-based approaches, e.g., 1 optimization and SPICE, with an infinitely dense grid. We generalize AST (atomic-norm soft thresholding) to the case of nonconsecutively sampled data (incomplete data) inspired by recent atomic norm based techniques. We present a gridless version… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
183
0

Year Published

2015
2015
2020
2020

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 271 publications
(184 citation statements)
references
References 54 publications
(125 reference statements)
1
183
0
Order By: Relevance
“…That is, the true signal DOAs may not necessarily fall on the exact discrete grid which is defined over a finite number of spatial angles. Off-grid sources cause so-called dictionary mismatch problem which violates the sparsity conditions and compromises the performance as well as the identifiability of the CS-based methods [19,20,21,22,23]. One solution to this problem is to use a denser search grid with an increased number of angles, thus leading to higher computational complexity.…”
Section: Introductionmentioning
confidence: 99%
“…That is, the true signal DOAs may not necessarily fall on the exact discrete grid which is defined over a finite number of spatial angles. Off-grid sources cause so-called dictionary mismatch problem which violates the sparsity conditions and compromises the performance as well as the identifiability of the CS-based methods [19,20,21,22,23]. One solution to this problem is to use a denser search grid with an increased number of angles, thus leading to higher computational complexity.…”
Section: Introductionmentioning
confidence: 99%
“…From a solution (X 11 , X 12 , X 22 ) of (4), we have r = rank(X 11 ) and the frequencies of the atoms a k = (1/ √ n)z n (ω k ) may be extracted by various methods, see e.g. [9,11,14].…”
Section: Sparse Optimizationmentioning
confidence: 99%
“…We know that the atomic set of ANM/RANM takes a continuous frequency value and can be regarded as a continuous dictionary that is the limiting scenario of the discretized dictionary in the GB-SR methods [10]. Without a discretized scheme, ANM/RANM can avoid the off-grid effect.…”
Section: A Anmmentioning
confidence: 99%
“…The GLS [10] method, which is a sparse and parametric method adopting the same covariance fitting criterion as SPICE [11] but without a discretizing scheme, is free of the off-grid effect and user parameters. According to (2), we suppose that the source signals and noise are uncorrelated.…”
Section: B Glsmentioning
confidence: 99%
See 1 more Smart Citation