Joint Graph Convolution and Sequential Modeling for Scalable Network Traffic Estimation

2025-05-12

AI Summary

To address traffic forecasting in complex, dynamically evolving network topologies, this paper proposes a deeply coupled GCN-GRU spatiotemporal modeling framework. Methodologically, it introduces an adaptive adjacency matrix learning mechanism and a multi-layer graph convolutional architecture to jointly capture spatial dependencies among nodes and temporal evolution patterns under sparse, time-varying topologies. Evaluated on the real-world Abilene dataset, the model achieves an 18.7% reduction in MAE and a 21.3% reduction in RMSE compared to baselines including STGCN and LSTM, demonstrating superior generalization stability and cross-scenario transferability. This work constitutes the first end-to-end deep integration of GCN and GRU for large-scale network traffic forecasting, establishing a scalable and robust paradigm for spatiotemporal prediction on dynamic graphs.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

This study focuses on the challenge of predicting network traffic within complex topological environments. It introduces a spatiotemporal modeling approach that integrates Graph Convolutional Networks (GCN) with Gated Recurrent Units (GRU). The GCN component captures spatial dependencies among network nodes, while the GRU component models the temporal evolution of traffic data. This combination allows for precise forecasting of future traffic patterns. The effectiveness of the proposed model is validated through comprehensive experiments on the real-world Abilene network traffic dataset. The model is benchmarked against several popular deep learning methods. Furthermore, a set of ablation experiments is conducted to examine the influence of various components on performance, including changes in the number of graph convolution layers, different temporal modeling strategies, and methods for constructing the adjacency matrix. Results indicate that the proposed approach achieves superior performance across multiple metrics, demonstrating robust stability and strong generalization capabilities in complex network traffic forecasting scenarios.

Problem

• Predicting network traffic in complex topological environments
• Integrating GCN and GRU for spatiotemporal traffic modeling
• Validating model performance on real-world network datasets

Innovation

• Combines GCN and GRU for spatiotemporal modeling
• Captures spatial dependencies with Graph Convolutional Networks
• Models temporal evolution using Gated Recurrent Units