ExposNet: A Deep Learning Framework for EMF Exposure Prediction in Complex Urban Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yarui, Wang, Shanshan, Wiart, Joe
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909525307555840
author Zhang, Yarui
Wang, Shanshan
Wiart, Joe
author_facet Zhang, Yarui
Wang, Shanshan
Wiart, Joe
contents The prediction of the electric field (E-field) plays a crucial role in monitoring radiofrequency electromagnetic field (RF-EMF) exposure induced by cellular networks. In this paper, a deep learning framework is proposed to predict E-field levels in complex urban environments. First, the measurement campaign and publicly accessible databases used to construct the training dataset are introduced, with a detailed explanation provided on how these datasets are formulated and integrated to enhance their suitability for Convolutional Neural Networks (CNNs)-based models. Then, the proposed model, ExposNet, is presented, and its network architecture and workflow are thoroughly explained. Two variations of the network structure are proposed, and extensive experimental analyses are conducted, demonstrating that ExposNet achieves good prediction accuracy with both configurations. Furthermore, the generalization capability of the model is evaluated. The overall results indicate that, despite being trained and tested on real-world measurements, the model performs well and achieves better accuracy compared to previous studies.
format Preprint
id arxiv_https___arxiv_org_abs_2503_02966
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ExposNet: A Deep Learning Framework for EMF Exposure Prediction in Complex Urban Environments
Zhang, Yarui
Wang, Shanshan
Wiart, Joe
Signal Processing
The prediction of the electric field (E-field) plays a crucial role in monitoring radiofrequency electromagnetic field (RF-EMF) exposure induced by cellular networks. In this paper, a deep learning framework is proposed to predict E-field levels in complex urban environments. First, the measurement campaign and publicly accessible databases used to construct the training dataset are introduced, with a detailed explanation provided on how these datasets are formulated and integrated to enhance their suitability for Convolutional Neural Networks (CNNs)-based models. Then, the proposed model, ExposNet, is presented, and its network architecture and workflow are thoroughly explained. Two variations of the network structure are proposed, and extensive experimental analyses are conducted, demonstrating that ExposNet achieves good prediction accuracy with both configurations. Furthermore, the generalization capability of the model is evaluated. The overall results indicate that, despite being trained and tested on real-world measurements, the model performs well and achieves better accuracy compared to previous studies.
title ExposNet: A Deep Learning Framework for EMF Exposure Prediction in Complex Urban Environments
topic Signal Processing
url https://arxiv.org/abs/2503.02966