Separable Autoregressive Moving Average Graph-Temporal Filters

E. Isufi; A. Loukas; A. Simonetto; G. Leus

doi:10.1109/eusipco.2016.7760238

Separable Autoregressive Moving Average Graph-Temporal Filters

E. Isufi, A. Loukas, A. Simonetto, G. Leus

Signal Processing Systems

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

Abstract

Despite their widespread use for the analysis of graph data, current graph filters are designed for graph signals that do not change over time, and thus they cannot simultaneously process time and graph frequency content in an adequate manner. This work presents ARMA2D, an autoregressive moving average graph-temporal filter that captures jointly the signal variations over the graph and time. By its unique nature, this filter is able to achieve a separable 2-dimensional frequency response, making it possible to approximate the filtering specifications along both the graph and temporal frequency domains. Numerical results show that the proposed solution outperforms the state of the art graph filters when the graph signal is time-varying.

Original language	English
Title of host publication	2016 24th European Signal Processing Conference (EUSIPCO)
Place of Publication	Piscataway
Publisher	IEEE
Pages	200-204
Number of pages	5
ISBN (Electronic)	978-0-9928-6265-7
ISBN (Print)	978-1-5090-1891-8
DOIs	https://doi.org/10.1109/eusipco.2016.7760238
Publication status	Published - 1 Dec 2016
Event	EUSIPCO 2016: 24th European Signal Processing Conference - Budapest, Hungary Duration: 29 Aug 2016 → 2 Sept 2016 Conference number: 24 http://www.eusipco2016.org/

Conference

Conference	EUSIPCO 2016
Abbreviated title	EUSIPCO
Country/Territory	Hungary
City	Budapest
Period	29/08/16 → 2/09/16
Internet address	http://www.eusipco2016.org/

Keywords

signal processing over graphs
graph filters
separable graph-temporal filters
distributed signal processing

Access to Document

10.1109/eusipco.2016.7760238

Cite this

@inproceedings{171f0a508b0c4c4f8e525e5287f6bf34,

title = "Separable Autoregressive Moving Average Graph-Temporal Filters",

abstract = "Despite their widespread use for the analysis of graph data, current graph filters are designed for graph signals that do not change over time, and thus they cannot simultaneously process time and graph frequency content in an adequate manner. This work presents ARMA2D, an autoregressive moving average graph-temporal filter that captures jointly the signal variations over the graph and time. By its unique nature, this filter is able to achieve a separable 2-dimensional frequency response, making it possible to approximate the filtering specifications along both the graph and temporal frequency domains. Numerical results show that the proposed solution outperforms the state of the art graph filters when the graph signal is time-varying.",

keywords = "signal processing over graphs, graph filters, separable graph-temporal filters, distributed signal processing",

author = "E. Isufi and A. Loukas and A. Simonetto and G. Leus",

year = "2016",

month = dec,

day = "1",

doi = "10.1109/eusipco.2016.7760238",

language = "English",

isbn = "978-1-5090-1891-8 ",

pages = "200--204",

booktitle = "2016 24th European Signal Processing Conference (EUSIPCO)",

publisher = "IEEE",

address = "United States",

note = "EUSIPCO 2016 : 24th European Signal Processing Conference, EUSIPCO ; Conference date: 29-08-2016 Through 02-09-2016",

url = "http://www.eusipco2016.org/",

}

TY - GEN

T1 - Separable Autoregressive Moving Average Graph-Temporal Filters

AU - Isufi, E.

AU - Loukas, A.

AU - Simonetto, A.

AU - Leus, G.

N1 - Conference code: 24

PY - 2016/12/1

Y1 - 2016/12/1

N2 - Despite their widespread use for the analysis of graph data, current graph filters are designed for graph signals that do not change over time, and thus they cannot simultaneously process time and graph frequency content in an adequate manner. This work presents ARMA2D, an autoregressive moving average graph-temporal filter that captures jointly the signal variations over the graph and time. By its unique nature, this filter is able to achieve a separable 2-dimensional frequency response, making it possible to approximate the filtering specifications along both the graph and temporal frequency domains. Numerical results show that the proposed solution outperforms the state of the art graph filters when the graph signal is time-varying.

AB - Despite their widespread use for the analysis of graph data, current graph filters are designed for graph signals that do not change over time, and thus they cannot simultaneously process time and graph frequency content in an adequate manner. This work presents ARMA2D, an autoregressive moving average graph-temporal filter that captures jointly the signal variations over the graph and time. By its unique nature, this filter is able to achieve a separable 2-dimensional frequency response, making it possible to approximate the filtering specifications along both the graph and temporal frequency domains. Numerical results show that the proposed solution outperforms the state of the art graph filters when the graph signal is time-varying.

KW - signal processing over graphs

KW - graph filters

KW - separable graph-temporal filters

KW - distributed signal processing

U2 - 10.1109/eusipco.2016.7760238

DO - 10.1109/eusipco.2016.7760238

M3 - Conference contribution

SN - 978-1-5090-1891-8

SP - 200

EP - 204

BT - 2016 24th European Signal Processing Conference (EUSIPCO)

PB - IEEE

CY - Piscataway

T2 - EUSIPCO 2016

Y2 - 29 August 2016 through 2 September 2016

ER -

Separable Autoregressive Moving Average Graph-Temporal Filters

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this