Sparse Bayesian learning with uncertain sensing matrix

Santosh Nannuru; Peter Gerstoft; Kay L. Gemba

ICASSP 2017accepted0 citations

Sparse Bayesian learning with uncertain sensing matrix

Santosh Nannuru, Peter Gerstoft, Kay L. Gemba

Abstract

Sparse Bayesian learning is a sparse processing method used for solving high-dimensional, underdetermined linear equations. Often the sensing matrix in the system of equations is assumed known and in presence of perturbations in this matrix performance of sparse processing degrades. We develop a sparse Bayesian learning method that accounts for perturbations in the sensing matrix. We derive an iterative weight update by performing evidence maximization. Beamforming simulations are used to demonstrate the advantages of the proposed method.

BibTeX

@inproceedings{icassp2017_sparsebayesianle,
  title = {Sparse Bayesian learning with uncertain sensing matrix},
  author = {Santosh Nannuru and Peter Gerstoft and Kay L. Gemba},
  booktitle = {ICASSP 2017},
  year = {2017}
}