Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals

Paolo Di Lorenzo; Elvin Isufi; Paolo Banelli; Sergio Barbarossa; Geert Leus

doi:10.23919/EUSIPCO.2017.8081618

Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals

Paolo Di Lorenzo, Elvin Isufi, Paolo Banelli, Sergio Barbarossa, Geert Leus

Signal Processing Systems

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

14 Citations (Scopus)

Abstract

This work proposes distributed recursive least squares (RLS) strategies for adaptive reconstruction and learning of signals defined over graphs. First, we introduce a centralized RLS estimation strategy with probabilistic sampling, and we propose a sparse sensing method that selects the sampling probability at each node in the graph in order to guarantee adaptive signal reconstruction and a target steady-state performance. Then, a distributed RLS strategy is derived and is shown to be convergent to its centralized counterpart. The performed numerical tests show the performance of the proposed adaptive method for distributed learning of graph signals.

Original language	English
Title of host publication	25th European Signal Processing Conference, EUSIPCO 2017
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	2289-2293
Number of pages	5
ISBN (Electronic)	978-0-9928626-7-1
DOIs	https://doi.org/10.23919/EUSIPCO.2017.8081618
Publication status	Published - 2017
Event	EUSIPCO 2017: 25th European Signal Processing Conference - Kos Island, Greece Duration: 28 Aug 2017 → 2 Sept 2017 Conference number: 25 https://www.eusipco2017.org/

Conference

Conference	EUSIPCO 2017
Abbreviated title	EUSIPCO
Country/Territory	Greece
City	Kos Island
Period	28/08/17 → 2/09/17
Internet address	https://www.eusipco2017.org/

Keywords

Recursive least squares estimation
graph signal processing
sampling
adaptive networks

Access to Document

10.23919/EUSIPCO.2017.8081618

Cite this

@inproceedings{5c4ad87a36fa4cd99f6b6ba3600c3e50,

title = "Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals",

abstract = "This work proposes distributed recursive least squares (RLS) strategies for adaptive reconstruction and learning of signals defined over graphs. First, we introduce a centralized RLS estimation strategy with probabilistic sampling, and we propose a sparse sensing method that selects the sampling probability at each node in the graph in order to guarantee adaptive signal reconstruction and a target steady-state performance. Then, a distributed RLS strategy is derived and is shown to be convergent to its centralized counterpart. The performed numerical tests show the performance of the proposed adaptive method for distributed learning of graph signals.",

keywords = "Recursive least squares estimation, graph signal processing, sampling, adaptive networks",

author = "{Di Lorenzo}, Paolo and Elvin Isufi and Paolo Banelli and Sergio Barbarossa and Geert Leus",

year = "2017",

doi = "10.23919/EUSIPCO.2017.8081618",

language = "English",

pages = "2289--2293",

booktitle = "25th European Signal Processing Conference, EUSIPCO 2017",

publisher = "IEEE",

address = "United States",

note = "EUSIPCO 2017 : 25th European Signal Processing Conference, EUSIPCO ; Conference date: 28-08-2017 Through 02-09-2017",

url = "https://www.eusipco2017.org/",

}

Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals. / Di Lorenzo, Paolo; Isufi, Elvin; Banelli, Paolo et al.
25th European Signal Processing Conference, EUSIPCO 2017. Piscataway, NJ: IEEE, 2017. p. 2289-2293.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals

AU - Di Lorenzo, Paolo

AU - Isufi, Elvin

AU - Banelli, Paolo

AU - Barbarossa, Sergio

AU - Leus, Geert

N1 - Conference code: 25

PY - 2017

Y1 - 2017

N2 - This work proposes distributed recursive least squares (RLS) strategies for adaptive reconstruction and learning of signals defined over graphs. First, we introduce a centralized RLS estimation strategy with probabilistic sampling, and we propose a sparse sensing method that selects the sampling probability at each node in the graph in order to guarantee adaptive signal reconstruction and a target steady-state performance. Then, a distributed RLS strategy is derived and is shown to be convergent to its centralized counterpart. The performed numerical tests show the performance of the proposed adaptive method for distributed learning of graph signals.

AB - This work proposes distributed recursive least squares (RLS) strategies for adaptive reconstruction and learning of signals defined over graphs. First, we introduce a centralized RLS estimation strategy with probabilistic sampling, and we propose a sparse sensing method that selects the sampling probability at each node in the graph in order to guarantee adaptive signal reconstruction and a target steady-state performance. Then, a distributed RLS strategy is derived and is shown to be convergent to its centralized counterpart. The performed numerical tests show the performance of the proposed adaptive method for distributed learning of graph signals.

KW - Recursive least squares estimation

KW - graph signal processing

KW - sampling

KW - adaptive networks

U2 - 10.23919/EUSIPCO.2017.8081618

DO - 10.23919/EUSIPCO.2017.8081618

M3 - Conference contribution

SP - 2289

EP - 2293

BT - 25th European Signal Processing Conference, EUSIPCO 2017

PB - IEEE

CY - Piscataway, NJ

T2 - EUSIPCO 2017

Y2 - 28 August 2017 through 2 September 2017

ER -

Distributed Recursive Least Squares Strategies for Adaptive Reconstruction of Graph Signals

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this