Forecasting Graph Signals with Recursive MIMO Graph Filters

Conference Paper (2023)
Author(s)

Jelmer van der Hoeven (Student TU Delft)

Alberto Natali (TU Delft - Signal Processing Systems)

Geert Leus (TU Delft - Signal Processing Systems)

DOI related publication
https://doi.org/10.23919/EUSIPCO58844.2023.10289997 Final published version
More Info
expand_more
Publication Year
2023
Language
English
Pages (from-to)
1843-1847
ISBN (print)
979-8-3503-2811-0
ISBN (electronic)
978-9-4645-9360-0
Event
Downloads counter
233
Collections
Institutional Repository
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Forecasting time series on graphs is a fundamental problem in graph signal processing. When each entity of the network carries a vector of values for each time stamp instead of a scalar one, existing approaches resort to the use of product graphs to combine this multidimensional information, at the expense of creating a larger graph. In this paper, we show the limitations of such approaches, and propose extensions to tackle them. Then, we propose a recursive multiple-input multiple-output graph filter which encompasses many already existing models in the literature while being more flexible. Numerical simulations on a real world data set show the effectiveness of the proposed models.

Files

Forecasting_Graph_Signals_with... (pdf)
(pdf | 0.344 Mb)
- Embargo expired in 01-05-2024
License info not available