2016
DOI: 10.1109/tsp.2016.2546231
|View full text |Cite
|
Sign up to set email alerts
|

Type I and Type II Bayesian Methods for Sparse Signal Recovery Using Scale Mixtures

Abstract: Abstract-In this paper, we propose a generalized scale mixture family of distributions, namely the Power Exponential Scale Mixture (PESM) family, to model the sparsity inducing priors currently in use for sparse signal recovery (SSR). We show that the successful and popular methods such as LASSO, Reweighted 1 and Reweighted 2 methods can be formulated in an unified manner in a maximum a posteriori (MAP) or Type I Bayesian framework using an appropriate member of the PESM family as the sparsity inducing prior. … Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

4
96
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 120 publications
(100 citation statements)
references
References 37 publications
4
96
0
Order By: Relevance
“…17,20 The hierarchical formulation of SBL inference offers both a computationally convenient Gaussian posterior distribution for adaptive processing (type-I maximum likelihood) and automatic regularization towards robust sparse estimates determined by the hyperparameters which maximize the evidence (type-II maximum likelihood). 21 In array signal processing, SBL is shown to improve significantly the resolution in beamforming 22 and in general the accuracy of DOA estimation, [23][24][25][26][27][28] 26 and multifrequency 23,24,27,28 SBL inference exploits the common sparsity profile across snapshots for stationary signals and frequencies for broadband signals to provide robust estimates by alleviating the ambiguity in the spatial mapping between sources and sensors due to noise and frequency-dependent spatial aliasing, respectively. Accounting for the statistics of modelling errors in SBL estimation, e.g., due to sensor position, sound speed uncertainty or basis mismatch, further improves support recovery.…”
mentioning
confidence: 99%
“…17,20 The hierarchical formulation of SBL inference offers both a computationally convenient Gaussian posterior distribution for adaptive processing (type-I maximum likelihood) and automatic regularization towards robust sparse estimates determined by the hyperparameters which maximize the evidence (type-II maximum likelihood). 21 In array signal processing, SBL is shown to improve significantly the resolution in beamforming 22 and in general the accuracy of DOA estimation, [23][24][25][26][27][28] 26 and multifrequency 23,24,27,28 SBL inference exploits the common sparsity profile across snapshots for stationary signals and frequencies for broadband signals to provide robust estimates by alleviating the ambiguity in the spatial mapping between sources and sensors due to noise and frequency-dependent spatial aliasing, respectively. Accounting for the statistics of modelling errors in SBL estimation, e.g., due to sensor position, sound speed uncertainty or basis mismatch, further improves support recovery.…”
mentioning
confidence: 99%
“…There has been increased interest in Bayesian probabilistic approaches to sparse signal recovery (SSR) problems, [4,20,21,29,32]. There are two categories of Bayesian probabilistic methods for SSR that encompass many well-known recovery algorithms in practice, [20]. The first is type-I, or maximum a posteriori (MAP) Bayesian estimation which uses a fixed prior.…”
Section: Probabilistic Approachmentioning
confidence: 99%
“…Capturing the sparsity of solutions more accurately in the SSR problem has been widely studied, [4,8,9,10,20,21,29,32]. More recently, there has been increased interest in Bayesian probabilistic approaches to SSR, [4,20,21,29,32]. Within the probabilistic approaches there are two categories.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Cramer-Rao bounds for SBL solution are discussed in [20]. Various sparse signal recovery solutions including LASSO and SBL are unified within the Bayesian framework in [21].…”
Section: Introductionmentioning
confidence: 99%