In this paper, we analyze a new class of iterative re-weighted least squares (IRLS) algorithms and their effectiveness in signal recovery from incomplete and inaccurate linear measurements. These methods can be interpreted as the constrained maximum likelihood estimation under a two-state Gaussian scale mixture assumption on the signal. We show that this class of algorithms, which performs exact recovery in noiseless scenarios under suitable assumptions, is robust even in presence of noise. Moreover these methods outperform classical IRLS for l_tau-minimization with tau in (0; 1] in terms of accuracy and rate of convergence.

Fast and Robust EM-Based IRLS Algorithm for Sparse Signal Recovery from Noisy Measurements / Ravazzi, Chiara; Magli, Enrico. - (2015), pp. 3841-3845. (Intervento presentato al convegno 40th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) nel Apr. 2015) [10.1109/ICASSP.2015.7178690].

Fast and Robust EM-Based IRLS Algorithm for Sparse Signal Recovery from Noisy Measurements

RAVAZZI, CHIARA;MAGLI, ENRICO
2015

Abstract

In this paper, we analyze a new class of iterative re-weighted least squares (IRLS) algorithms and their effectiveness in signal recovery from incomplete and inaccurate linear measurements. These methods can be interpreted as the constrained maximum likelihood estimation under a two-state Gaussian scale mixture assumption on the signal. We show that this class of algorithms, which performs exact recovery in noiseless scenarios under suitable assumptions, is robust even in presence of noise. Moreover these methods outperform classical IRLS for l_tau-minimization with tau in (0; 1] in terms of accuracy and rate of convergence.
File in questo prodotto:
File Dimensione Formato  
FullText_Ravazzi_ICASSP2015.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 235.09 kB
Formato Adobe PDF
235.09 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2592665
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo