Saved in:
Bibliographic Details
Main Authors: Yan, Zijiang, Huang, Yixiang, Pei, Jianhua, Tabassum, Hina, Chiaraviglio, Luca
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.15067
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913131398168576
author Yan, Zijiang
Huang, Yixiang
Pei, Jianhua
Tabassum, Hina
Chiaraviglio, Luca
author_facet Yan, Zijiang
Huang, Yixiang
Pei, Jianhua
Tabassum, Hina
Chiaraviglio, Luca
contents The rapid growth in wireless infrastructure has increased the need to accurately estimate and forecast electromagnetic field (EMF) levels to ensure ongoing compliance, assess potential health impacts, and support efficient network planning. While existing studies rely on univariate forecasting of wideband aggregate EMF data, frequency-selective multivariate forecasting is needed to capture the inter-operator and inter-frequency variations essential for proactive network planning. To this end, this paper introduces EMFusion, a conditional multivariate diffusion-based probabilistic forecasting framework that integrates diverse contextual factors, such as time of day, season, and holidays, while providing explicit uncertainty estimates. The proposed architecture features a residual U-Net backbone enhanced by a cross-attention mechanism that dynamically integrates external conditions to guide the generation process. Furthermore, EMFusion integrates an imputation-based sampling strategy that treats forecasting as a structural inpainting task, ensuring temporal coherence even with irregular measurements. Unlike standard point forecasters, EMFusion generates empirical probabilistic prediction intervals from the learned conditional distribution, providing uncertainty-aware probabilistic forecasting rather than simple point estimation. Numerical experiments conducted on frequency-selective EMF datasets demonstrate that EMFusion with the contextual information of working hours outperforms the baseline models with or without conditions. EMFusion outperforms the best baseline by 23.85% in continuous ranked probability score (CRPS), 13.93% in normalized root mean square error, and reduces prediction CRPS error by 22.47%.
format Preprint
id arxiv_https___arxiv_org_abs_2512_15067
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EMFusion: An Uncertainty-Aware Conditional Diffusion Framework for Frequency-Selective EMF Forecasting in Wireless Networks
Yan, Zijiang
Huang, Yixiang
Pei, Jianhua
Tabassum, Hina
Chiaraviglio, Luca
Machine Learning
Artificial Intelligence
Systems and Control
The rapid growth in wireless infrastructure has increased the need to accurately estimate and forecast electromagnetic field (EMF) levels to ensure ongoing compliance, assess potential health impacts, and support efficient network planning. While existing studies rely on univariate forecasting of wideband aggregate EMF data, frequency-selective multivariate forecasting is needed to capture the inter-operator and inter-frequency variations essential for proactive network planning. To this end, this paper introduces EMFusion, a conditional multivariate diffusion-based probabilistic forecasting framework that integrates diverse contextual factors, such as time of day, season, and holidays, while providing explicit uncertainty estimates. The proposed architecture features a residual U-Net backbone enhanced by a cross-attention mechanism that dynamically integrates external conditions to guide the generation process. Furthermore, EMFusion integrates an imputation-based sampling strategy that treats forecasting as a structural inpainting task, ensuring temporal coherence even with irregular measurements. Unlike standard point forecasters, EMFusion generates empirical probabilistic prediction intervals from the learned conditional distribution, providing uncertainty-aware probabilistic forecasting rather than simple point estimation. Numerical experiments conducted on frequency-selective EMF datasets demonstrate that EMFusion with the contextual information of working hours outperforms the baseline models with or without conditions. EMFusion outperforms the best baseline by 23.85% in continuous ranked probability score (CRPS), 13.93% in normalized root mean square error, and reduces prediction CRPS error by 22.47%.
title EMFusion: An Uncertainty-Aware Conditional Diffusion Framework for Frequency-Selective EMF Forecasting in Wireless Networks
topic Machine Learning
Artificial Intelligence
Systems and Control
url https://arxiv.org/abs/2512.15067