DF-3DRME: A Data-Friendly Learning Framework for 3D Radio Map Estimation based on Super-Resolution Technique

2026-04-01

AI Summary

This work addresses the high cost and labor-intensive data acquisition typically required for constructing high-resolution three-dimensional radio maps. To mitigate this challenge, the authors propose DF-3DRME, a novel framework that introduces, for the first time in this domain, a data-efficient hybrid supervised learning paradigm. By leveraging a small number of high-resolution samples alongside abundant low-resolution measurements, DF-3DRME employs a two-stage network architecture: an LR-Net first generates coarse radio maps, which are subsequently refined via an SR-Net performing super-resolution reconstruction. Remarkably, the method achieves excellent reconstruction performance with high-resolution labels available for only 4% of the environment, drastically reducing reliance on costly high-resolution annotations. This substantial decrease in data requirements significantly lowers deployment costs and enhances practical feasibility for real-world applications.

Abstract

High-Resolution three-dimensional (3D) radio maps (RMs) provide rich information about the radio landscape that is essential to a myriad of wireless applications in the future wireless networks. Although deep learning (DL) methods have shown their effectiveness in RM construction, existing approaches require massive high-resolution 3D RM samples in the training dataset, the acquisition of which is labor-intensive and time-consuming in practice. In this paper, our goal is to devise a data-friendly high-resolution 3D RM construction solution via training over a hybrid dataset, wherein the RMs associated with a small fraction of environment maps (EMs) are of high-resolution, while those corresponding to the majority of EMs are of low-resolution. To this end, we propose a Data-Friendly 3D Radio Map Estimator (DF-3DRME), which comprises two processing stages. Specifically, in the first stage, we leverage the abundant low-resolution 3D RM samples to train a neural network, termed the LR-Net, for predicting the low-resolution 3D RM from the input EM, which provides a coarse characterization of the spatial radio propagation. In the second stage, we employ an advanced super-resolution network, termed the SR-Net, to upscale the predicted low-resolution 3D RM to its high-resolution counterpart. Unlike the LR-Net, the SR-Net can be effectively trained with only the limited high-resolution 3D RM samples available in the hybrid dataset. Experimental results demonstrate that the proposed framework achieves compelling reconstruction performance with only 4% of the EMs in the dataset having high-resolution 3D RM labels, which significantly reduces data acquisition overhead and facilitates practical deployment.

Problem

• 3D radio map estimation
• data-efficient learning
• super-resolution
• hybrid dataset
• wireless networks

Innovation

• 3D radio map estimation
• data-friendly learning
• super-resolution
• hybrid-resolution dataset
• deep learning