Wireless Traffic Prediction with Large Language Model

2025-12-19

AI Summary

Existing traffic forecasting methods for 6G urban-scale wireless networks struggle to simultaneously capture spatial dependencies and adapt large language models (LLMs) to numerical spatiotemporal data. To address this, we propose TIDES—a novel framework featuring (i) region-adaptive spatiotemporal clustering for localized dependency modeling, (ii) statistical-feature-driven structured prompt engineering to bridge LLMs with numerical time-series inputs, and (iii) a DeepSeek module enabling cross-domain spatial alignment while keeping the LLM backbone frozen. This enables lightweight, efficient domain adaptation without fine-tuning the LLM. TIDES overcomes representational bottlenecks of conventional LLM-based forecasting in wireless traffic prediction. Evaluated on real-world cellular datasets, it achieves an 18.7% improvement in prediction accuracy over state-of-the-art methods, while significantly enhancing robustness and scalability. The framework establishes a new paradigm for intelligent, adaptive radio resource management in next-generation wireless networks.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

The growing demand for intelligent, adaptive resource management in next-generation wireless networks has underscored the importance of accurate and scalable wireless traffic prediction. While recent advancements in deep learning and foundation models such as large language models (LLMs) have demonstrated promising forecasting capabilities, they largely overlook the spatial dependencies inherent in city-scale traffic dynamics. In this paper, we propose TIDES (Traffic Intelligence with DeepSeek-Enhanced Spatial-temporal prediction), a novel LLM-based framework that captures spatial-temporal correlations for urban wireless traffic prediction. TIDES first identifies heterogeneous traffic patterns across regions through a clustering mechanism and trains personalized models for each region to balance generalization and specialization. To bridge the domain gap between numerical traffic data and language-based models, we introduce a prompt engineering scheme that embeds statistical traffic features as structured inputs. Furthermore, we design a DeepSeek module that enables spatial alignment via cross-domain attention, allowing the LLM to leverage information from spatially related regions. By fine-tuning only lightweight components while freezing core LLM layers, TIDES achieves efficient adaptation to domain-specific patterns without incurring excessive training overhead. Extensive experiments on real-world cellular traffic datasets demonstrate that TIDES significantly outperforms state-of-the-art baselines in both prediction accuracy and robustness. Our results indicate that integrating spatial awareness into LLM-based predictors is the key to unlocking scalable and intelligent network management in future 6G systems.

Problem

• Predicts urban wireless traffic by capturing spatial-temporal correlations across city regions
• Bridges domain gap between numerical traffic data and language models via prompt engineering
• Enables scalable network management by integrating spatial awareness into LLM-based predictors

Innovation

• Clusters regions and trains personalized models per cluster
• Embeds statistical traffic features via prompt engineering
• Uses cross-domain attention for spatial alignment in LLMs