Modelling indoor electromagnetic fields (EMF) from mobile phone base stations for epidemiological studies

Area of Science:

• Environmental Health
• Radio Frequency Engineering
• Epidemiological Methods

Background:

• Accurate assessment of indoor radio frequency electromagnetic fields (RF-EMF) from mobile base stations is vital for epidemiological studies, as individuals spend most time indoors.
• Existing 3D radio wave propagation models are effective for outdoor RF-EMF but require validation for indoor environments.
• Understanding factors influencing indoor RF-EMF exposure is essential for reliable health risk assessments.

Purpose of the Study:

• To evaluate the accuracy of indoor RF-EMF model predictions from mobile base stations.
• To determine if incorporating building characteristics improves indoor RF-EMF model accuracy.
• To assess the feasibility of ranking indoor RF-EMF exposure for epidemiological research.

Main Methods:

• Conducted 15-minute spot measurements of RF-EMF in 263 rooms across 101 primary schools and 30 homes in Amsterdam.
• Collected data on building characteristics, including glazing and wall/window frame materials.
• Utilized the NISMap 3D radio wave propagation model to predict RF-EMF and compared predictions with measurements.
• Employed mixed-effects models to explore the influence of building characteristics on model-measurement agreement.

Main Results:

• A Spearman correlation of 0.73 was observed between modelled and measured total downlink RF-EMF from base stations.
• Average modelled RF-EMF was 0.053 mW/m², and average measured RF-EMF was 0.041 mW/m².
• The precision of the model was 0.184 mW/m²; incorporating building characteristics did not significantly improve prediction accuracy.
• Despite some exposure misclassification, the model demonstrated feasibility in ranking indoor RF-EMF levels.

Conclusions:

• The NISMap model provides a feasible method for ranking indoor RF-EMF exposure from mobile base stations, which is valuable for epidemiological studies.
• Building characteristics, such as glazing and material types, did not enhance the accuracy of the studied indoor RF-EMF propagation model.
• Further research may be needed to refine models for more precise individual indoor RF-EMF exposure assessment, but current models support relative exposure ranking.