Distributed sparsified graph filters for denoising and diffusion tasks

Elvin Isufi; Geert Leus

doi:10.1109/ICASSP.2017.7953281

Distributed sparsified graph filters for denoising and diffusion tasks

^*Corresponding author for this work

Signal Processing Systems

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

7 Citations (Scopus)

Abstract

Generally in distributed signal processing, and specifically in distributed graph filters, reducing the communication and computational complexity plays a key role in the network lifetime. In this work we present a novel algorithm to sparsify the graph filtering operation in a random way, where each node decides locally with a certain probability with which of its neighbors to communicate. We show that, if the filter coefficients are changed accordingly, the first and second order moment of the stochastic output are identical to the deterministic filter output and bounded, respectively. We apply our idea on the tasks of signal denoising and diffusion. Numerical results show that the distributed implementation costs of the filter can be reduced up to 95% with a variance of 10-3 from the deterministic output.

Original language	English
Title of host publication	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	5865-5869
Number of pages	5
ISBN (Electronic)	978-1-5090-4117-6
DOIs	https://doi.org/10.1109/ICASSP.2017.7953281
Publication status	Published - 2017
Event	ICASSP 2017: 42nd IEEE International Conference on Acoustics, Speech and Signal Processing - The Internet of Signals - Hilton New Orleans Riverside, New Orleans, LA, United States Duration: 5 Mar 2017 → 9 Mar 2017 Conference number: 42 http://www.ieee-icassp2017.org/

Conference

Conference	ICASSP 2017
Abbreviated title	ICASSP
Country/Territory	United States
City	New Orleans, LA
Period	5/03/17 → 9/03/17
Internet address	http://www.ieee-icassp2017.org/

Keywords

diffusion graph signals
graph filters
graph signal denoising
graph signal processing
graph sparsification

Access to Document

10.1109/ICASSP.2017.7953281

Cite this

@inproceedings{23192f397e1e4e69b27ef82f2952862c,

title = "Distributed sparsified graph filters for denoising and diffusion tasks",

abstract = "Generally in distributed signal processing, and specifically in distributed graph filters, reducing the communication and computational complexity plays a key role in the network lifetime. In this work we present a novel algorithm to sparsify the graph filtering operation in a random way, where each node decides locally with a certain probability with which of its neighbors to communicate. We show that, if the filter coefficients are changed accordingly, the first and second order moment of the stochastic output are identical to the deterministic filter output and bounded, respectively. We apply our idea on the tasks of signal denoising and diffusion. Numerical results show that the distributed implementation costs of the filter can be reduced up to 95% with a variance of 10-3 from the deterministic output.",

keywords = "diffusion graph signals, graph filters, graph signal denoising, graph signal processing, graph sparsification",

author = "Elvin Isufi and Geert Leus",

year = "2017",

doi = "10.1109/ICASSP.2017.7953281",

language = "English",

pages = "5865--5869",

booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings",

publisher = "IEEE",

address = "United States",

note = "ICASSP 2017 : 42nd IEEE International Conference on Acoustics, Speech and Signal Processing - The Internet of Signals, ICASSP ; Conference date: 05-03-2017 Through 09-03-2017",

url = "http://www.ieee-icassp2017.org/",

}

Distributed sparsified graph filters for denoising and diffusion tasks. / Isufi, Elvin ; Leus, Geert.
2017 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. Piscataway, NJ: IEEE, 2017. p. 5865-5869 7953281.

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

TY - GEN

T1 - Distributed sparsified graph filters for denoising and diffusion tasks

AU - Isufi, Elvin

AU - Leus, Geert

N1 - Conference code: 42

PY - 2017

Y1 - 2017

N2 - Generally in distributed signal processing, and specifically in distributed graph filters, reducing the communication and computational complexity plays a key role in the network lifetime. In this work we present a novel algorithm to sparsify the graph filtering operation in a random way, where each node decides locally with a certain probability with which of its neighbors to communicate. We show that, if the filter coefficients are changed accordingly, the first and second order moment of the stochastic output are identical to the deterministic filter output and bounded, respectively. We apply our idea on the tasks of signal denoising and diffusion. Numerical results show that the distributed implementation costs of the filter can be reduced up to 95% with a variance of 10-3 from the deterministic output.

AB - Generally in distributed signal processing, and specifically in distributed graph filters, reducing the communication and computational complexity plays a key role in the network lifetime. In this work we present a novel algorithm to sparsify the graph filtering operation in a random way, where each node decides locally with a certain probability with which of its neighbors to communicate. We show that, if the filter coefficients are changed accordingly, the first and second order moment of the stochastic output are identical to the deterministic filter output and bounded, respectively. We apply our idea on the tasks of signal denoising and diffusion. Numerical results show that the distributed implementation costs of the filter can be reduced up to 95% with a variance of 10-3 from the deterministic output.

KW - diffusion graph signals

KW - graph filters

KW - graph signal denoising

KW - graph signal processing

KW - graph sparsification

UR - http://www.scopus.com/inward/record.url?scp=85023778475&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7953281

DO - 10.1109/ICASSP.2017.7953281

M3 - Conference contribution

AN - SCOPUS:85023778475

SP - 5865

EP - 5869

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings

PB - IEEE

CY - Piscataway, NJ

T2 - ICASSP 2017

Y2 - 5 March 2017 through 9 March 2017

ER -

Distributed sparsified graph filters for denoising and diffusion tasks

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this