An Exposimetric Electromagnetic Comparison of Mobile Phone Emissions: 5G versus 4G Signals Analyses by Means of Statistics and Convolutional Neural Networks Classification
Abstract
Overview
To examine the non-thermal biological effects of microwaves, employing both new metrics and methodologies becomes essential. This research indicates a necessity to shift from time-averaged analysis towards peak exposure analysis. Unlike previous studies, this research does not compare the general characteristics of 4G and 5G mobile communication signals. Instead, it focuses on real-life exposure conditions varying across several parameters.
Findings
- Methodology: Utilized signal and spectrum analyzers which include statistical analysis using amplitude probability density (APD), complementary cumulative distribution function (CCDF), channel power measurements, and spectrogram databases.
- Results: Analyzed various mobile phone usage scenarios such as file downloading, uploading, video streaming, and video calls across 4G and 5G networks. Highlighted the significant distinction in amplitude-time features based on the type of network and application.
- Electric Fields: Noted the highest and lowest electric field strengths at a 10 cm distance from the mobile device during different operations. Also, detected and classified the signal emissions using a convolutional neural network with remarkable accuracy.
Conclusion
The study introduces a novel metric for assessing the dynamic human exposure to EMF through various mobile applications. It underlines the crucial role of real-time and peak analysis over traditional averaged data, enabling a deeper understanding of the exposure implications from mobile communications, particularly 5G over 4G technology. Highlights the necessity for ongoing research into these exposures under varied real-world conditions.