The balanced weighted orthogonal matching pursuit (bWOMP) algorithm for recovering signals in compressed sensing (CS) based system is presented as a specialized recovering tool for Electrocardiograph (ECG) signals. Being based on the standard OMP approach, bWOMP is a lightweight reconstruction algorithm both in terms of complexity and memory footprint. Furthermore, the concept of weighting is introduced in the algorithm by exploring a prior knowledge on ECG signals. Experimental results show a performance increase of about 10 dB with respect to the standard OMP approach, and also an increase with respect to the decoding approaches considered as the state-of-the-art. In this case the gain could be as high as 4 dB with respect to the best of currently known decoding approaches.

Low-complexity greedy algorithm in compressed sensing for the adapted decoding of ECGs / Marchioni, A.; Mangia, M.; Pareschi, F.; Rovatti, R.; Setti, G.. - STAMPA. - 2018:(2017), pp. 1-4. (Intervento presentato al convegno 2017 IEEE Biomedical Circuits and Systems Conference, BioCAS 2017 tenutosi a Torino (Italy) nel October 19-21, 2017) [10.1109/BIOCAS.2017.8325143].

Low-complexity greedy algorithm in compressed sensing for the adapted decoding of ECGs

Pareschi F.;Setti G.
2017

Abstract

The balanced weighted orthogonal matching pursuit (bWOMP) algorithm for recovering signals in compressed sensing (CS) based system is presented as a specialized recovering tool for Electrocardiograph (ECG) signals. Being based on the standard OMP approach, bWOMP is a lightweight reconstruction algorithm both in terms of complexity and memory footprint. Furthermore, the concept of weighting is introduced in the algorithm by exploring a prior knowledge on ECG signals. Experimental results show a performance increase of about 10 dB with respect to the standard OMP approach, and also an increase with respect to the decoding approaches considered as the state-of-the-art. In this case the gain could be as high as 4 dB with respect to the best of currently known decoding approaches.
2017
978-1-5090-5803-7
File in questo prodotto:
File Dimensione Formato  
08325143.pdf

non disponibili

Descrizione: Editorial Version
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 183.54 kB
Formato Adobe PDF
183.54 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
BioCAS2017-WOMP.pdf

accesso aperto

Descrizione: Author version of the Paper
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 362.92 kB
Formato Adobe PDF
362.92 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/11583/2786597