Study on field strength prediction using different models on time series from urban continuous RF-EMF monitoring

Authors: Song X, Feng W, Yang C, Djuric N, Kljajic D, Djuric S

Year: 2025

Category: Environmental Science

Journal: Expert Systems with Applications

DOI: 10.1016/j.eswa.2025.126963

URL: https://www.sciencedirect.com/science/article/pii/S0957417425005858

Abstract

Overview

Continuous and intensive monitoring of electromagnetic fields (EMFs), particularly in urban areas, is crucial due to prolonged human exposure. These areas are notable for increased sensitivity to radio-frequency (RF) EMF levels.

Findings

The study examines the efficacy of various predictive models on RF-EMF strength in urban settings, focusing on Novi Sad, Serbia. The models analyzed include SARIMA, CNN, LSTM, ELM, PLS, and Transformers. This research pinpoints areas like kindergartens and schools as critical zones for EMF exposure.

Data Utilized: Two years of RF-EMF data
Comparative Metrics: Prediction accuracy, performance degradation, extreme value accuracy, and training duration

Conclusion

The PLS model surpasses others in predicting EMF exposure levels. This study advocates for its application in real-time monitoring systems to bolster public health protection and spur further research in targeted EMF management.

EMF monitoring RF-EMF prediction urban environments public health machine learning model comparison electromagnetic field

← Back to Stats