GOLIAT team develops new modelling method to assess 5G exposure

Frame of the 3D digital model of the city of Ghent used to test the new method

In the framework of Project GOLIAT, a team of researchers from Ghent University has developed a new method to estimate exposure to electromagnetic fields (EMFs) in 5G networks in different urban environments. The results of their work have been published in IEEE Access.

Until now, models for estimating exposure to EMFs have been inefficient for assessing human exposure in 5G networks due to the peculiarities of this technology. In 5G networks, base stations direct signals to specific users, a technique called “beamforming”. As a result, EMFs don’t remain stable in time or space and depend on the user’s location and behaviour. To respond to these challenges, the new model developed by the GOLIAT team takes into account both beamforming and people’s mobility.

Since in the real world, people living in cities are constantly moving, the method integrates an agent-based model (ABM) that allows for the simulation of pedestrian movement in a realistic city-wide environment. In addition, it incorporates a technique called ray-tracing to simulate how electromagnetic waves propagate in an urban scenario. The new model also offers the possibility to simulate different antenna configurations and precoding schemes. 

Precoding is the manipulation of signals before transmission to optimise their performance. In this study, they ran simulations with two of the most common precoding techniques: zero-forcing (ZF) and maximum ratio transmission (MRT).

All results well below the threshold

To validate the method, the researchers used maps, terrain data and mobile phone base station information from Belgium to create a realistic 3D urban environment in which to run their simulations.

The simulation results showed that a higher number of antenna elements resulted in higher field strenghts for active phone users, underscoring the increased directionality of larger arrays towards users. Despite this increased directionality, a larger number of antenna elements (antenna arrays) also meant a higher exposure for people who weren’t actively using their phones, something that the authors attribute to their spatial proximity to active users. A higher number of active users resulted in lower field strengths, because the power is distributed among all of them. In terms of precoding schemes, it was found that the zero-forcing scheme resulted in a lower exposure for both people who were using their phones and people who weren’t. 

Most importantly, the results showed that even in the settings where the highest antenna power was simulated, exposure levels to electromagnetic fields levels was less than 4% of the threshold set by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) for the general population. 

“Our findings provide critical insights into the interaction between antenna configuration, precoding and user dynamics, offering a novel perspective on exposure modelling in realistic 5G environments”, says Matthias Leeman, researcher at Ghent University and first author of the study. 

“Our model opens an avenue for a more realistic investigation of human exposure to 5G”, says Wout Joseph, researcher at Ghent University and last author of the study. 

Reference

M. Leeman, R. Wydaeghe, J. van der Straeten, S. Goegebeur, G. Vermeeren and W. Joseph, “City-Scale Spatio-Temporal Modeling of 5G Downlink Exposure of Users and Non-Users by Ray-Tracing in a Real Urban Environment,” in IEEE Access, vol. 13, pp. 30894-30906, 2025, doi: 10.1109/ACCESS.2025.3541352