Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting

With rapid urbanization and technological advancements, increasing population density has posed significant challenges for urban traffic management. Traffic flow forecasting is essential for optimizing traffic signals, regulating traffic flow, and improving overall transportation efficiency. Recent...

Full description

Saved in:
Bibliographic Details
Main Authors: Liyuan Wang, Jiafeng Zhuang, Shuo Ma, Hai Lin
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11036171/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850114115083698176
author Liyuan Wang
Jiafeng Zhuang
Shuo Ma
Hai Lin
author_facet Liyuan Wang
Jiafeng Zhuang
Shuo Ma
Hai Lin
author_sort Liyuan Wang
collection DOAJ
description With rapid urbanization and technological advancements, increasing population density has posed significant challenges for urban traffic management. Traffic flow forecasting is essential for optimizing traffic signals, regulating traffic flow, and improving overall transportation efficiency. Recent advancements in deep learning, particularly Graph Convolutional Networks (GCNs), offer more effective solutions for traffic forecasting. However, existing approaches face challenges in modeling traffic trends, periodicity, and randomness. To address these issues, this paper proposes a Frequency-enhanced dynamic Graph Convolutional Network (FGCN) that incorporates sequence decomposition and reconstruction to extract periodic and trend components. Additionally, the proposed method integrates the Fourier transform into the loss function to mitigate label autocorrelation and improve long-term forecasting accuracy. Furthermore, this paper employs a dynamic graph convolution method to separately model spatial and temporal dependencies, capturing traffic flow variations more effectively. Experimental results on four open datasets demonstrate that FGCN not only outperforms state-of-the-art methods but also provides interpretable insights into the dynamic spatial relationships between road segments, while significantly improving forecasting accuracy.
format Article
id doaj-art-50d4cce679764b0ab63e18f8e6fe96a2
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-50d4cce679764b0ab63e18f8e6fe96a22025-08-20T02:36:59ZengIEEEIEEE Access2169-35362025-01-011310345110346210.1109/ACCESS.2025.357933411036171Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow ForecastingLiyuan Wang0Jiafeng Zhuang1Shuo Ma2https://orcid.org/0009-0003-3665-9739Hai Lin3https://orcid.org/0000-0003-1495-7121CCCC Second Highway Consultants Company Ltd., Wuhan, ChinaCCCC Second Highway Consultants Company Ltd., Wuhan, ChinaSchool of Cyber Science and Engineering, Wuhan University, Wuhan, ChinaSchool of Cyber Science and Engineering, Wuhan University, Wuhan, ChinaWith rapid urbanization and technological advancements, increasing population density has posed significant challenges for urban traffic management. Traffic flow forecasting is essential for optimizing traffic signals, regulating traffic flow, and improving overall transportation efficiency. Recent advancements in deep learning, particularly Graph Convolutional Networks (GCNs), offer more effective solutions for traffic forecasting. However, existing approaches face challenges in modeling traffic trends, periodicity, and randomness. To address these issues, this paper proposes a Frequency-enhanced dynamic Graph Convolutional Network (FGCN) that incorporates sequence decomposition and reconstruction to extract periodic and trend components. Additionally, the proposed method integrates the Fourier transform into the loss function to mitigate label autocorrelation and improve long-term forecasting accuracy. Furthermore, this paper employs a dynamic graph convolution method to separately model spatial and temporal dependencies, capturing traffic flow variations more effectively. Experimental results on four open datasets demonstrate that FGCN not only outperforms state-of-the-art methods but also provides interpretable insights into the dynamic spatial relationships between road segments, while significantly improving forecasting accuracy.https://ieeexplore.ieee.org/document/11036171/Traffic flow forecastingdeep learningspatiotemporal modelinggraph convolutional networks
spellingShingle Liyuan Wang
Jiafeng Zhuang
Shuo Ma
Hai Lin
Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting
IEEE Access
Traffic flow forecasting
deep learning
spatiotemporal modeling
graph convolutional networks
title Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting
title_full Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting
title_fullStr Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting
title_full_unstemmed Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting
title_short Frequency Enhanced Dynamic Graph Convolutional Networks for Traffic Flow Forecasting
title_sort frequency enhanced dynamic graph convolutional networks for traffic flow forecasting
topic Traffic flow forecasting
deep learning
spatiotemporal modeling
graph convolutional networks
url https://ieeexplore.ieee.org/document/11036171/
work_keys_str_mv AT liyuanwang frequencyenhanceddynamicgraphconvolutionalnetworksfortrafficflowforecasting
AT jiafengzhuang frequencyenhanceddynamicgraphconvolutionalnetworksfortrafficflowforecasting
AT shuoma frequencyenhanceddynamicgraphconvolutionalnetworksfortrafficflowforecasting
AT hailin frequencyenhanceddynamicgraphconvolutionalnetworksfortrafficflowforecasting