Instrumentation & Measurement Magazine 23-9 - 5

On the Measurement of Human
Exposure to Cellular Networks
Giovanni Betta, Domenico Capriglione, Gianni Cerro, Gianfranco Miele,
Marzia Salone D'Amata, Darko Šuka, Predrag Pejovic´, and Mirjana Simic´-Pejovic´

T

he measurement of Human Exposure to Electromagnetic Fields (EMFs) is an important topic in
today's scenarios. Indeed, the even more pervasive
diffusion of cellular networks (also with incoming 5G technology), as well as short-range and personal wireless devices,
has increased interest in human exposure issues. Focusing the
attention on cellular systems, although regulators and normative committees are conveying specific guidelines for the
measurements of EMFs generated by second-generation (2G),
third-generation (3G) and fourth-generation (4G) mobile
communication systems, there are still several issues to be addressed regarding the most reliable measurement procedures
and evaluation of the related measurement uncertainties. In
particular, the high variability of voice and data traffic makes it
very difficult to achieve reliable forecasting of human exposure
to 3G and 4G systems in both narrowband and dosimetric measurements. Therefore, refinement of measurement methods
and procedures, as well as data post-processing and averaging
techniques, are still open issues that must be investigated in detail by the measurement and networking community.

Open Issues in EMF Monitoring
The evaluation of human exposure to electric, magnetic and
electromagnetic fields is a topic of great interest today, specifically related to Radio Frequency (RF) EMFs generated by
telecommunication systems. In the last decades, there has
been even more diffusion of sources of EMFs designed for
several purposes, such as: short-range communications (as
Bluetooth or ZigBee); Wireless Local Area Networks, i.e., Wi-Fi
in its several releases; Wireless Sensor Networks, radio-modem technologies, and communication protocols, for ad-hoc
point-to-point networks or metering applications; long-range
radio; TV communications; and cellular networks, conveying
services like voice and data at very high data‑rate. All cited
technologies coexist today, especially in the 100 kHz to 3 GHz
band, and therefore great attention is paid to this frequency
range from the human exposure point of view. Although a
clear correlation between Non-Ionizing RF fields' human exposure and possible health risks is not proved, a precautionary
December 2020	

approach is followed worldwide to protect people from possible effects due to long‑term exposure to EMFs. To this aim,
suitable limits have been defined by each country, following
the International Commission on Non-Ionizing Radio Protection (ICNIRP) guidelines and the World Health Organization
(WHO) recommendations [1], [2].
As for cellular systems, the last few years showed great efforts in the development of technical standards to describe
measurement instruments, techniques and procedures to be
applied for the evaluation of human exposure related to such
systems. Scientific literature proposed both narrowband and
broadband measurement techniques and procedures aimed
at minimizing the measurement chain uncertainty [3], [4].
However, besides the instruments' metrological features,
the overall uncertainty is also affected by the traffic variability during the day. In [5] and [6], it has been proved how the
variability of the emissions generated by a cellular base station during the day and on different days can be thought of
as either bias and/or measurement uncertainty components.
These contributions could make the evaluation of the actual
human exposure unpredictable or unreliable.
As for the narrowband measurements, it is generally possible to perform selective measurements to estimate the
overall field contribution due to each source and frequency
band. Recently, extrapolation techniques for maximal exposure estimation due to cellular systems have been adopted by
international standards [4] because they look efficient in the estimation of the maximum electric field strength (EFS) that a
Base Station (BS) can emit. Consequently, if the estimated EFS
is lower than the applicable limit, likely, such a limit will not
be overcome during the typical BS operating cycle. Indeed,
the extrapolation techniques assume that control and/or signaling channels' power levels are constant, thus making the
measurement day (in the year) and interval (in the day) choices
irrelevant. In summary, such techniques are thought to keep
the measurement time low, thus making a frequency selective
measurement campaign practical, especially when the evaluation of the EFS due to each operator, cellular technology and
frequency channel for the considered BS must be carried out.

IEEE Instrumentation & Measurement Magazine	5
1094-6969/20/$25.00©2020IEEE
Instrumentation & Measurement Magazine 23-9

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 23-9
