LSDM: LLM-Enhanced Spatio-temporal Diffusion Model for Service-Level Mobile Traffic Prediction

2025-07-23

AI Summary

To address poor adaptability, high uncertainty, insufficient environmental context awareness, and challenges in modeling cross-service dependencies for service-level mobile traffic forecasting in urban environments, this paper proposes an LLM-enhanced Spatio-Temporal Diffusion Model (LLM-SDM). The method integrates a large language model (LLM) into the diffusion process to explicitly encode multi-source urban semantic features—including points of interest (POIs), road networks, and weather—while leveraging a Transformer architecture to jointly model spatio-temporal dynamics and multimodal contextual dependencies. Extensive experiments on real-world mobile traffic datasets demonstrate that LLM-SDM significantly improves forecasting accuracy: the coefficient of determination (R²) increases by at least 2.83% over baseline methods, and the root mean square error (RMSE) decreases by at least 8.29% compared to state-of-the-art models such as CSDI. The approach further exhibits superior generalization capability and robustness under varying urban conditions.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Diffusion Models for Vision

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Service-level mobile traffic prediction for individual users is essential for network efficiency and quality of service enhancement. However, current prediction methods are limited in their adaptability across different urban environments and produce inaccurate results due to the high uncertainty in personal traffic patterns, the lack of detailed environmental context, and the complex dependencies among different network services. These challenges demand advanced modeling techniques that can capture dynamic traffic distributions and rich environmental features. Inspired by the recent success of diffusion models in distribution modeling and Large Language Models (LLMs) in contextual understanding, we propose an LLM-Enhanced Spatio-temporal Diffusion Model (LSDM). LSDM integrates the generative power of diffusion models with the adaptive learning capabilities of transformers, augmented by the ability to capture multimodal environmental information for modeling service-level patterns and dynamics. Extensive evaluations on real-world service-level datasets demonstrate that the model excels in traffic usage predictions, showing outstanding generalization and adaptability. After incorporating contextual information via LLM, the performance improves by at least 2.83% in terms of the coefficient of determination. Compared to models of a similar type, such as CSDI, the root mean squared error can be reduced by at least 8.29%. The code and dataset will be available at: https://github.com/SoftYuaneR/LSDM.

Problem

• Predicting service-level mobile traffic for individual users accurately
• Capturing dynamic traffic distributions and environmental features effectively
• Enhancing adaptability across diverse urban environments and services

Innovation

• LLM-enhanced diffusion model for traffic prediction
• Integrates diffusion models with transformer capabilities
• Captures multimodal environmental information dynamically