EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Frequency-Guided Dynamic Hypergraph Learning for Traffic Flow Forecasting

Wanqi Li iD, Bin Wang, Gang Li, Yan Ma iD, Botao Jiang

Sensors · 19.08.2026 · 26 (16) · 5238

DOI: 10.3390/s26165238 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 19.08.2026. Verfasst von Wanqi Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate traffic flow forecasting requires modeling both stable macroscopic dependencies and abrupt local fluctuations in complex road networks. Existing spatiotemporal forecasting models usually learn spatial structures from raw time-domain traffic signals, where low-frequency trends and high-frequency fluctuations are entangled. Although decomposition-based and frequency-aware methods have shown the benefit of separating heterogeneous traffic components, how frequency decomposition can support reliable high-order topology learning remains less explored. To address this issue, we propose FEDHNet, a Frequency-Guided Dynamic Hypergraph Network for traffic flow forecasting. FEDHNet first performs adaptive spectral decomposition on the hidden representation to obtain low-frequency and complementary high-frequency latent components. The low-frequency branch constructs dynamic hyperedges from the relatively smooth latent representation to model non-local high-order dependencies, while the high-frequency branch employs a lightweight 2D Inception module with GLU-based gated denoising to model rapidly varying latent responses. A low-frequency-anchored residual fusion module then adaptively integrates high-frequency residual information into the low-frequency latent representation for multi-step prediction. Experiments on four public PeMS datasets show that FEDHNet achieves competitive forecasting accuracy and multi-horizon performance, together with favorable computational efficiency compared with recent spatiotemporal forecasting baselines. Further analyses examine the effects of topology-source selection and controlled high-frequency residual modeling, revealing that the benefit of low-frequency hypergraph construction is dataset-dependent.

Schlagwörter

Stichwörter der Autor:innen

dynamic hypergraph learningfrequency decompositionspatio-temporal graph neural networkstraffic flow forecasting

Autor:innen und Institutionen

  • College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, China.
  • Zhongke Zidong Information Technology (Beijing) Co., Ltd., Beijing 100190, China.
  • China Satellite Network Application Co., Ltd., Beijing 100000, China.

ORCID-Profile: Wanqi Li, Yan Ma

Zitieren

APA

Li, W., Wang, B., Li, G., Ma, Y., & Jiang, B. (2026). Frequency-Guided Dynamic Hypergraph Learning for Traffic Flow Forecasting. Sensors, 26(16), 5238. https://doi.org/10.3390/s26165238

Vancouver

Li W, Wang B, Li G, Ma Y, Jiang B. Frequency-Guided Dynamic Hypergraph Learning for Traffic Flow Forecasting. Sensors. 2026;26(16):5238. doi:10.3390/s26165238

BibTeX
@article{Li2026,
  author = {Li, Wanqi and Wang, Bin and Li, Gang and Ma, Yan and Jiang, Botao},
  title = {Frequency-Guided Dynamic Hypergraph Learning for Traffic Flow Forecasting},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {16},
  pages = {5238},
  doi = {10.3390/s26165238}
}
Kostenlos starten →