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

TL;DR

This paper introduces LSDM, a novel model combining diffusion models and LLMs to improve service-level mobile traffic prediction by capturing complex dependencies and environmental factors, demonstrating superior accuracy and adaptability.

Contribution

The paper presents a new LLM-Enhanced Spatio-temporal Diffusion Model that integrates diffusion models with transformers and environmental context for better traffic prediction.

Findings

Performance improves by at least 2.83% with contextual information.

Root mean squared error reduced by at least 8.29% compared to similar models.

Model shows strong generalization and adaptability on real-world datasets.

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.