EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multiscale Traffic Dynamics Representation for Forecasting via MEMD-Guided Dual-Branch Recurrent Networks

Yichen Qian iD, Taiming Kang iD, Shengduo Zhang iD, Chaoneng Li iD, Xiaolong Wang iD, Shuxu Zhao iD

Sensors · 26.05.2026 · 26 (11) · 3369

DOI: 10.3390/s26113369 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 26.05.2026. Verfasst von Yichen Qian und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Gansu Provincial Science and Technology Department, Gansu Education Department, Joint Innovation Fund Project of Lanzhou Jiaotong University and Tianjin University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traffic flow forecasting remains challenging because raw traffic flow observations often contain mixed temporal patterns, including slowly varying trends and fast local fluctuations. To address this issue, this paper proposes a Multivariate Empirical Mode Decomposition (MEMD)-guided dual-branch recurrent framework for multistep point forecasting. Specifically, MEMD is used as an alignment-preserving multivariate decomposition mechanism to obtain frequency-aligned components, which are then reconstructed into low-frequency trend and high-frequency residual components. The trend component is modeled by a Long Short-Term Memory (LSTM) branch to capture smooth long-term evolution, while the residual component is learned by a Bidirectional Gated Recurrent Unit (Bi-GRU) branch to characterize short-term oscillatory dynamics. A lightweight fusion head is then used to integrate the two branch-specific representations for final prediction. Experiments on PeMS04 and PeMS08, two traffic datasets derived from the California Department of Transportation Performance Measurement System, show that the proposed method achieves competitive performance across mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE), reaching 19.67/31.59/12.95% on PeMS04 and 15.51/24.43/9.86% on PeMS08. Compared with representative recent baselines, the proposed method achieves competitive results, with relative gains reaching 5.89% on PeMS04 and 5.35% on PeMS08 in selected metric-wise comparisons. These results indicate that MEMD-guided trend–residual representation learning can improve multistep traffic flow forecasting.

Schlagwörter

Stichwörter der Autor:innen

dual-branch recurrent networkmultiscale temporal modelingmultivariate empirical mode decomposition (MEMD)traffic flow forecasting

Autor:innen und Institutionen

  • School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China.

ORCID-Profile: Yichen Qian, Taiming Kang, Shengduo Zhang, Chaoneng Li, Xiaolong Wang, Shuxu Zhao

Förderung

  • Gansu Provincial Science and Technology Department
  • Gansu Education Department
  • Joint Innovation Fund Project of Lanzhou Jiaotong University and Tianjin University
  • Gansu Provincial Department of Education University Faculty Innovation Fund Project

Zitieren

APA

Qian, Y., Kang, T., Zhang, S., Li, C., Wang, X., & Zhao, S. (2026). Multiscale Traffic Dynamics Representation for Forecasting via MEMD-Guided Dual-Branch Recurrent Networks. Sensors, 26(11), 3369. https://doi.org/10.3390/s26113369

Vancouver

Qian Y, Kang T, Zhang S, Li C, Wang X, Zhao S. Multiscale Traffic Dynamics Representation for Forecasting via MEMD-Guided Dual-Branch Recurrent Networks. Sensors. 2026;26(11):3369. doi:10.3390/s26113369

BibTeX
@article{Qian2026,
  author = {Qian, Yichen and Kang, Taiming and Zhang, Shengduo and Li, Chaoneng and Wang, Xiaolong and Zhao, Shuxu},
  title = {Multiscale Traffic Dynamics Representation for Forecasting via MEMD-Guided Dual-Branch Recurrent Networks},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {11},
  pages = {3369},
  doi = {10.3390/s26113369}
}
Kostenlos starten →