Mobile Traffic Prediction using LLMs with Efficient In-context Demonstration Selection

TL;DR

This paper presents a novel LLM-based framework for mobile traffic prediction that uses a two-step demonstration selection strategy to improve accuracy, validated on real 5G data with superior results over existing methods.

Contribution

It introduces a two-step demonstration selection strategy for LLMs in mobile traffic prediction, enhancing accuracy through effectiveness and informativeness rules.

Findings

Lower mean squared error compared to baseline methods

Higher R2-Scores than zero-shot and other demonstration selection techniques

Effective on real-world 5G datasets across different scenarios

Abstract

Mobile traffic prediction is an important enabler for optimizing resource allocation and improving energy efficiency in mobile wireless networks. Building on the advanced contextual understanding and generative capabilities of large language models (LLMs), this work introduces a context-aware wireless traffic prediction framework powered by LLMs. To further enhance prediction accuracy, we leverage in-context learning (ICL) and develop a novel two-step demonstration selection strategy, optimizing the performance of LLM-based predictions. The initial step involves selecting ICL demonstrations using the effectiveness rule, followed by a second step that determines whether the chosen demonstrations should be utilized, based on the informativeness rule. We also provide an analytical framework for both informativeness and effectiveness rules. The effectiveness of the proposed framework is demonstrated with a real-world fifth-generation (5G) dataset with different application scenarios. According to the numerical results, the proposed framework shows lower mean squared error and higher R2-Scores compared to the zero-shot prediction method and other demonstration selection methods, such as constant ICL demonstration selection and distance-only-based ICL demonstration selection.