In this paper, we address the problem of in-network reconstruction of correlated sparse signals. Specifically, we adopt a JSM-1 model by which the signals to be reconstructed are the sum of a common sparse term and an individual sparse term (or innovation). We tackle the problem using an Alternating Direction Method of Multipliers approach, which is prone to be distributed. We also propose a version that requires to exchange only binary messages among neighboring nodes. Performance of the different methods is shown to be satisfactory.

Distributed ADMM for in-network reconstruction of sparse signals with innovations / J., Matamoros; Fosson, Sophie; Magli, Enrico; Carles Anton, Haro. - (2014), pp. 429-433. (Intervento presentato al convegno 2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP)) [10.1109/GlobalSIP.2014.7032153].

Distributed ADMM for in-network reconstruction of sparse signals with innovations

FOSSON, SOPHIE;MAGLI, ENRICO;
2014

Abstract

In this paper, we address the problem of in-network reconstruction of correlated sparse signals. Specifically, we adopt a JSM-1 model by which the signals to be reconstructed are the sum of a common sparse term and an individual sparse term (or innovation). We tackle the problem using an Alternating Direction Method of Multipliers approach, which is prone to be distributed. We also propose a version that requires to exchange only binary messages among neighboring nodes. Performance of the different methods is shown to be satisfactory.
File in questo prodotto:
File Dimensione Formato  
FullText-MatamorosGlobalSIP2014.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 265.11 kB
Formato Adobe PDF
265.11 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/11583/2592663
 Attenzione

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