Network traffic prediction based on feature fusion spatio-temporal graph convolutional network

Miaoyi Zhou; Yechen He; Wenyuan Li; Dong Liang

doi:10.1117/12.3032905

5 July 2024 Network traffic prediction based on feature fusion spatio-temporal graph convolutional network

Miaoyi Zhou, Yechen He, Wenyuan Li, Dong Liang

Proceedings Volume 13184, Third International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024); 131846K (2024) https://doi.org/10.1117/12.3032905
Event: 3rd International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024), 2024, Kuala Lumpur, Malaysia

Abstract

Network traffic prediction is a crucial aspect of network management and operation, attracting significant attention from both academic and industrial researchers. Deep learning-based approaches to network traffic prediction, particularly those employing spatio-temporal features, have seen continuous advancements. However, current models struggle with the heterogeneity and complex spatio-temporal dependencies of network data, often falling short in accurately reflecting realworld network traffic dynamics. To tackle this problem, in this paper we propose a novel Feature Fusion Spatio-Temporal Graph Convolutional Network named FFSTGCN for network traffic prediction, which aims to enhance spatio-temporal feature extraction and uncover latent node relationships more effectively. Specifically, we first present a novel feature fusion module to learn more features from network traffic data by constructing two adjacency matrices with different expressions. We then propose a new random walk mechanism that incorporate Gated Recurrent Unit (GRU) to capture potential paths and relationships between nodes in the network. To target the extraction of critical information from the network, we further design corresponding attention mechanisms in the temporal and spatial domains for performance enhancement. Experimental evaluations on public network traffic datasets demonstrate that our proposed FFSTGCN framework can achieve superior performance in network traffic prediction tasks.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Miaoyi Zhou, Yechen He, Wenyuan Li, and Dong Liang "Network traffic prediction based on feature fusion spatio-temporal graph convolutional network", Proc. SPIE 13184, Third International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024), 131846K (5 July 2024); https://doi.org/10.1117/12.3032905

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
14 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Data modeling

Feature fusion

Matrices

Performance modeling

Machine learning

Telecommunication networks

Feature extraction

Show All Keywords

Keywords/Phrases

Search In:

Publication Years