Title
Forecasting Graph Signals with Recursive MIMO Graph Filters
Author
van der Hoeven, Jelmer (Student TU Delft)
Natali, A. (TU Delft Signal Processing Systems)
Leus, G.J.T. (TU Delft Signal Processing Systems) 
Date
2023
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.
Subject
Forecasting
Graph Signal Processing
Product Graph
Multi-dimensional graph signals
To reference this document use:
http://resolver.tudelft.nl/uuid:5760cc99-d872-455e-ad16-992563ad617e
DOI
https://doi.org/10.23919/EUSIPCO58844.2023.10289997
Publisher
IEEE
Embargo date
2024-05-01
ISBN
979-8-3503-2811-0
Source
Proceedings of the 2023 31st European Signal Processing Conference (EUSIPCO)
Event
31st European Signal Processing Conference, 2023-09-04 → 2023-09-08, Helsinki, Finland
Series
European Signal Processing Conference, 2219-5491
Bibliographical note
Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Part of collection
Institutional Repository
Document type
conference paper
Rights
© 2023 Jelmer van der Hoeven, A. Natali, G.J.T. Leus