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

2025/06/05 by Han Zhang, Zhang, Han, Akram Bin Sediq +5
Computer Science · Engineering · Social Sciences · #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Network Security and Intrusion Detection #Networking and Internet Architecture (cs.NI) #Traffic Prediction and Management Techniques

paper · pdf · doi:10.48550/arxiv.2506.12074

openalex publication_date 2025/06/05 · openalex created_date 2025/10/13 · openalex updated_date 2026/07/28

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.

Citations

• LLM-Inference-Bench: Inference Benchmarking of Large Language Models on AI Accelerators
• Self-Refined Generative Foundation Models for Wireless Traffic Prediction
• Why does in-context learning fail sometimes? Evaluating in-context learning on open and closed questions
• LLM-Based Intent Processing and Network Optimization Using Attention-Based Hierarchical Reinforcement Learning
• Why Larger Language Models Do In-context Learning Differently?
• Large Language Models in Wireless Application Design: In-Context Learning-enhanced Automatic Network Intrusion Detection
• Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
• Transformer-Based Wireless Traffic Prediction and Network Optimization in O-RAN
• In-Context Learning with Iterative Demonstration Selection
• Large Language Models Are Zero-Shot Time Series Forecasters
• TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting
• Heuristic Algorithms for RIS-assisted Wireless Networks: Exploring Heuristic-aided Machine Learning
• RL meets Multi-Link Operation in IEEE 802.11be: Multi-Headed Recurrent Soft-Actor Critic-based Traffic Allocation
• One Fits All:Power General Time Series Analysis by Pretrained LM
• Large Language Models Can Be Easily Distracted by Irrelevant Context
• What Makes Good Examples for Visual In-Context Learning?
• Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Ordering
• Why Can GPT Learn In-Context? Language Models Implicitly Perform Gradient Descent as Meta-Optimizers
• SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
• Active Example Selection for In-Context Learning
• Phase Shift Compression for Control Signaling Reduction in IRS-Aided Wireless Systems: Global Attention and Lightweight Design
• Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?
• Learning To Retrieve Prompts for In-Context Learning
• An Explanation of In-context Learning as Implicit Bayesian Inference
• Realtime Mobile Bandwidth and Handoff Predictions in 4G/5G Networks
• Intelligent Reflecting Surface Assisted Terahertz Communications toward 6G
• The Road Towards 6G: A Comprehensive Survey
• CBAM: Convolutional Block Attention Module
• Understanding Emergent In-Context Learning from a Kernel Regression Perspective

Related

• Self-Refined Generative Foundation Models for Wireless Traffic Prediction
• Wireless Traffic Prediction with Large Language Model
• UoMo: A Universal Model of Mobile Traffic Forecasting for Wireless Network Optimization
• User Traffic Prediction for Proactive Resource Management:\n Learning-Powered Approaches
• Mobility, traffic and radio channel prediction: 5G and beyond applications