Mathematical Sciences

Lund University

Title Sparse Estimation Of Spectroscopic Signals
Authors Stefan Ingi Adalbjörnsson, Andreas Jakobsson
Alternative Location
Publication European Signal Processing Conference
Year 2011
Volume 2011
Pages 333 - 337
Document type Conference paper
Conference name 19th European Signal Processing Conference, EUSIPCO 2011
Conference Date 2011-08-29/2011-09-02
Conference Location Barcelona
Status Published
Quality controlled Yes
Language eng
Publisher European Association for Signal Processing (EURASIP)
Abstract English This work considers the semi-parametric estimation of sparse spec-<br> troscopic signals, aiming to form a detailed spectral representation<br> of both the frequency content and the spectral line widths of the oc-<br> curring signals. Extending on the recent FOCUSS-based SLIM al-<br> gorithm, we propose an alternative prior for a Bayesian formulation<br> of this sparse reconstruction method, exploiting a proposed suitable<br> prior for the noise variance. Examining three common models for<br> spectroscopic signals, the introduced technique allows for reliable<br> estimation of the characteristics of these models. Numerical sim-<br> ulations illustrate the improved performance of the proposed tech-<br> nique.
ISBN/ISSN/Other ISSN: 2219-5491 (online)