An Exposimetric Electromagnetic Comparison of Mobile Phone Emissions: 5G versus 4G Signals Analyses by Means of Statistics and Convolutional Neural Networks Classification

Abstract

Overview

This study explores the topic of non-thermal biological effects of microwaves, suggesting that new metrics and methodologies, such as peak exposure analysis, are vital for gaining a better understanding. The focus is not on comparing general characteristics between 4G and 5G mobile communications but on analyzing specific real-life exposure conditions which may vary widely.

Findings

• A detailed amplitude-time feature differentiation of the 4G vs. 5G signals is presented.
• We provided analyses based on types of mobile applications: file download, upload, internet video streaming, and video calls, identifying varying levels of electric field strengths.
• The electric field strengths reported show higher emissions for 5G, particularly highlighted during internet video streaming where 5G strength was three times higher than that of 4G.
• Differences in amplitude probability density distributions and crest factors between the two technologies were statistically significant, with 5G showing higher variance.
• The application of convolutional neural networks enabled high accuracy in classifying the emissions, enhancing understanding of exposure dynamics.

Conclusion

The study provides critical insights into the electromagnetic field (EMF) exposure variances between emerging 5G technology and existing 4G technology, underlining significant increases in EMF exposure that confirm ongoing health risk debates. A thorough investigation into these differences is crucial to better understand potential non-thermal biological effects on human health necessitating further academic scrutiny.