Radio Propagation Modelling: To Differentiate or To Deep Learn, That Is The Question

2025-09-15

AI Summary

Current radio propagation modeling lacks systematic empirical evidence on the scalability and practical efficacy of differentiable ray tracing versus deep learning models in production-grade wireless networks. This study conducts the first large-scale comparative evaluation across urban, suburban, and rural environments using over 10,000 real-world base stations across 13 cities. Methodologically, it systematically validates differentiable ray tracing at operator scale, revealing its inherent limitations in generalization and real-time performance. Concurrently, it demonstrates that deep learning models deliver significant advantages for coverage prediction: achieving an average accuracy gain of 3 dB, higher training and inference efficiency, and strong cross-scenario robustness. The results provide critical empirical evidence and actionable technical guidance for deploying wireless digital twins and enabling intelligent network optimization.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Differentiable ray tracing has recently challenged the status quo in radio propagation modelling and digital twinning. Promising unprecedented speed and the ability to learn from real-world data, it offers a real alternative to conventional deep learning (DL) models. However, no experimental evaluation on production-grade networks has yet validated its assumed scalability or practical benefits. This leaves mobile network operators (MNOs) and the research community without clear guidance on its applicability. In this paper, we fill this gap by employing both differentiable ray tracing and DL models to emulate radio coverage using extensive real-world data collected from the network of a major MNO, covering 13 cities and more than 10,000 antennas. Our results show that, while differentiable ray-tracing simulators have contributed to reducing the efficiency-accuracy gap, they struggle to generalize from real-world data at a large scale, and they remain unsuitable for real-time applications. In contrast, DL models demonstrate higher accuracy and faster adaptation than differentiable ray-tracing simulators across urban, suburban, and rural deployments, achieving accuracy gains of up to 3 dB. Our experimental results aim to provide timely insights into a fundamental open question with direct implications on the wireless ecosystem and future research.

Problem

• Evaluating differentiable ray tracing versus deep learning for radio propagation modeling
• Assessing scalability and practical benefits using real-world mobile network data
• Determining which approach better generalizes across diverse urban environments

Innovation

• Differentiable ray tracing for radio propagation modeling
• Deep learning models for radio coverage emulation
• Comparative evaluation using real-world mobile network data