Recovery of traffic information through graph signal processing

Abstract The generation of data sets from traffic variables within the roads of a city is increasing due to the implementation of sensors, monitoring stations, or more elaborate systems, such as synchronised drones to record the dynamics of a city. However, each method has a limited coverage area an...

Full description

Saved in:
Bibliographic Details
Main Authors: Rafael Alejandro Martínez Márquez, Giuseppe Patanè
Format: Article
Language:English
Published: SpringerOpen 2025-04-01
Series:EURASIP Journal on Advances in Signal Processing
Subjects:
Online Access:https://doi.org/10.1186/s13634-025-01211-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850265518263500800
author Rafael Alejandro Martínez Márquez
Giuseppe Patanè
author_facet Rafael Alejandro Martínez Márquez
Giuseppe Patanè
author_sort Rafael Alejandro Martínez Márquez
collection DOAJ
description Abstract The generation of data sets from traffic variables within the roads of a city is increasing due to the implementation of sensors, monitoring stations, or more elaborate systems, such as synchronised drones to record the dynamics of a city. However, each method has a limited coverage area and it is necessary to extrapolate the values for the road network. Constructing a road graph of a city, each node represents a road and the edges represent the intersection between two roads. Considering the values of the traffic variables as graph signals, we apply graph signal recovery methods based on distinct notions of smooth graph signals to complete the values on the whole network from the available values on a small subset of nodes. The experimental results on the pNEUMA data set on the Athens road graph show that the recovery methods are a potential solution to completing the traffic status on an urban network. Additionally, as one of the further interpretations that are possible to obtain from a complete graph traffic signal, we use the recovered volume signal to estimate the traffic flows within the area of Athens covered by its road graph through Fick’s law based on the combinatorial Laplacian matrix. We show that the smallest flow values are located in the zones with a higher number of points of interest, and also that residential areas are the main sources and destinations of traffic flows.
format Article
id doaj-art-d96b37f6f3c6456e8619a376d438eca8
institution OA Journals
issn 1687-6180
language English
publishDate 2025-04-01
publisher SpringerOpen
record_format Article
series EURASIP Journal on Advances in Signal Processing
spelling doaj-art-d96b37f6f3c6456e8619a376d438eca82025-08-20T01:54:25ZengSpringerOpenEURASIP Journal on Advances in Signal Processing1687-61802025-04-012025112610.1186/s13634-025-01211-0Recovery of traffic information through graph signal processingRafael Alejandro Martínez Márquez0Giuseppe Patanè1CNR-IMATI, Consiglio Nazionale delle RicercheCNR-IMATI, Consiglio Nazionale delle RicercheAbstract The generation of data sets from traffic variables within the roads of a city is increasing due to the implementation of sensors, monitoring stations, or more elaborate systems, such as synchronised drones to record the dynamics of a city. However, each method has a limited coverage area and it is necessary to extrapolate the values for the road network. Constructing a road graph of a city, each node represents a road and the edges represent the intersection between two roads. Considering the values of the traffic variables as graph signals, we apply graph signal recovery methods based on distinct notions of smooth graph signals to complete the values on the whole network from the available values on a small subset of nodes. The experimental results on the pNEUMA data set on the Athens road graph show that the recovery methods are a potential solution to completing the traffic status on an urban network. Additionally, as one of the further interpretations that are possible to obtain from a complete graph traffic signal, we use the recovered volume signal to estimate the traffic flows within the area of Athens covered by its road graph through Fick’s law based on the combinatorial Laplacian matrix. We show that the smallest flow values are located in the zones with a higher number of points of interest, and also that residential areas are the main sources and destinations of traffic flows.https://doi.org/10.1186/s13634-025-01211-0Traffic data recoveryGraph signal recoveryKernel ridge regressionRegularisation functionsApproximately bandlimited graph signals
spellingShingle Rafael Alejandro Martínez Márquez
Giuseppe Patanè
Recovery of traffic information through graph signal processing
EURASIP Journal on Advances in Signal Processing
Traffic data recovery
Graph signal recovery
Kernel ridge regression
Regularisation functions
Approximately bandlimited graph signals
title Recovery of traffic information through graph signal processing
title_full Recovery of traffic information through graph signal processing
title_fullStr Recovery of traffic information through graph signal processing
title_full_unstemmed Recovery of traffic information through graph signal processing
title_short Recovery of traffic information through graph signal processing
title_sort recovery of traffic information through graph signal processing
topic Traffic data recovery
Graph signal recovery
Kernel ridge regression
Regularisation functions
Approximately bandlimited graph signals
url https://doi.org/10.1186/s13634-025-01211-0
work_keys_str_mv AT rafaelalejandromartinezmarquez recoveryoftrafficinformationthroughgraphsignalprocessing
AT giuseppepatane recoveryoftrafficinformationthroughgraphsignalprocessing