Modelling of Total Exposure in Hypothetical 5G Mobile Networks for Varied Topologies and User Scenarios

2019 · Mixed · Low evidence

← Research stream

MixedLow evidencePaper 1538

Study snapshot

Publicationmanual · 2019

Study typeengineering

Reported directionMixed

Evidence ratingLow

Abstract

Modelling of Total Exposure in Hypothetical 5G Mobile Networks for Varied Topologies and User Scenarios Sven Kuehn, Serge Pfeifer, Beyhan Kochali, Niels Kuster. Modelling of Total Exposure in Hypothetical 5G Mobile Networks for Varied Topologies and User Scenarios. Final Report of Project CRR-816. A report on behalf of the Swiss Federal Office for the Environment (FOEN). IT'IS Foundation, Zurich. 24 June 2019. Executive Summary In January 2019, the Swiss Federal Office for the Environment (FOEN) mandated the IT'IS Foundation to evaluate the total human exposure in hypothetical 5G mobile networks for varied topologies and user scenarios to identify factors that would minimize the total exposure of the population. In this study, total exposure is defined as the combined exposure from network base stations, the user's own device, as well as bystanders' mobile devices. The influence of various factors on total exposure in mobile communication networks (as defined above) was modeled and analyzed with the help of the Monte Carlo simulation technique. Total exposure is described as the local peak specific absorption rate (SAR) spatially averaged over any 10 g of tissue mass (psaSAR10g) averaged over a period of 6 minutes. The unit psaSAR10g was chosen because it defines the governing basic restriction for wireless exposure as the whole-body average SAR limits (wbaSAR) are intrinsically met if the limits of local exposure are satisfied. The averaging duration of 6 minutes constitutes the internationally accepted averaging time to prevent thermal hazards at frequencies below 6 GHz as instant values have little justification. However, it should be noted that some regulators define shorter averaging time periods, e.g., the US Federal Communications Commission (FCC) of 100 s. In a first step, we analyzed the tissue-specific exposure as a function of frequency. The preliminary dosimetric study showed that exposure of the human brain to the 3.6 GHz band, that has been recently added to the Swiss mobile communication frequencies, is reduced by a factor of >6 for the tissue averaged SAR when compared to mobile network operation at <1 GHz. This reduction is due to the smaller penetration depth at higher frequencies. This conclusion, however, does not apply to exposed tissues close to the surface or skin (eyes, testicles, etc.) when the peak SAR in this tissue is evaluated. The peak SAR in the grey matter remains in approximately the same order of magnitude ( 3 dB) over all frequencies but the area of high exposure is reduced at 3.6 GHz. In a second step, we used data measured in 4G systems and analyzed the latest mobile network standards to extrapolate the exposures for various 5G network scenarios. These measured data were also used to extrapolate the exposure to the future development of data usage in 5G networks. Specifically, we analyzed the effect on the total exposure of (i) the network topology by varying the cell size and amount of indoor coverage in the network, as well as the usage of (ii) an individual's own device, and (iii) devices of close bystanders. The results - based on simulations of more than 200 different exposure scenarios - reveal that, for all user types, except for non-users (including passive mobile phone users and users dominantly using downlink data traffic, e.g., video streaming), total exposure is dominated by the person's own mobile device. Compared to non- users, the exposure is increased (i) for light users (with 100 MByte uplink data per day) by 6 - 10 dB (or a factor of 4 to 10), (ii) for moderate users (with 1 GByte uplink data per day) by 13 - 25 dB (or a factor of 20 to >300), and (iii) for heavy users by 15 - 40 dB (or a factor of 30 to >10000). Further, the results show that peak exposure of non-users is not defined by exposure to base stations but by exposure to mobile devices of close bystanders in urban areas resulting in 6 dB (or a factor of 4) higher exposure than from a nearby base station antenna. While a reduction of the mobile cell size leads to a reduction in total exposure by a factor of 2 to10 for people actively using their mobile devices, this might also lead to a small increase by a factor of 1.6 in total exposure of non-users due the generally increased incident signal levels from the surrounding base stations. Similarly, the exposure of active users can be reduced by a factor of 4 to 600 by increasing the indoor network coverage. Yet, in line with the results for the mobile cell sizes, increased indoor coverage will also lead to increased exposure of non-users by a factor of 2 to 10. This increase, however, starts at a level 1000 times lower than the typical total exposure of active users. The results of this study show that the personal mobile device is the dominant exposure source for active mobile network users. Besides a person's own usage behavior, total exposure is also closely linked to the network infrastructure. Generally speaking, a network with a lower path loss, i.e., smaller cells and additional indoor coverage, helps to reduce total exposure. The exposure per transmitted bit is reduced by a factor of <3 by the increased spectral efficiency of the 5G technology, and the reduced penetration depth associated with the new bands at 3.5 - 3.8 GHz. The results presented above are limited due to the network data that has been used and the definition of total exposure as stated in this report. Furthermore, it only considers time-averaged (6 min) and not instant exposures. This study does not consider (i) the effect of upcoming massive MIMO systems in 5G networks, (ii) alternative data transmission links, for instance the use of Wireless Local Area Network (WLAN), and (iii) millimeter wave frequencies in 5G mobile networks. Conclusions The results of this study show that the absorption of energy by the human brain, resulting from exposure to the 3.6 GHz band newly added to the Swiss mobile communication frequencies, isreduced by a factor >6 for the tissue averaged SAR when compared to mobile networks operating at <1 GHz, and by a factor of >2 when compared to the frequency bands at 1.8 - 2GHz. For deep brain regions, the reduction is much larger. The reduced exposure for these regions is due to lower penetration depths at higher frequencies. Close to the surface (eyes, testicles, etc.) the exposure can be higher. At the most exposed surface of the grey matter, the values remain approximately 3 dB over all frequencies whereas the area of high exposure is reduced. More than 200 Monte Carlo simulated exposure scenarios have been analyzed to evaluate total human exposure in 5G Networks for different topologies and user scenarios. The results show that for all users (except non-users), the total exposure is dominated by a person's own mobile device. Compared to a non-user, the exposure is increased for a light user (with 100 MByte uplink data per day) by 6 - 10 dB (or by a factor 4 to 10), for a moderate user (with 1 GByte uplink data per day) by 13 - 25 dB (or by a factor of 20 to >300), and for a heavy user by 25 - 40 dB (or a factor of 300 to >10000). The peak exposure of non-users is further not defined by exposure to surrounding base stations but by mobile devices of close bystanders in urban areas,resulting in 6 dB (or a factor of 4) higher exposure than from a nearby base station antenna. Reducing the diameter of the mobile cell leads to a decreased overall exposure by a factor of 2 to 10 for people who actively use their mobile devices. At the same time, the reduction in cell size might lead to a small increase by a factor <2 in exposure for non-users. The exposure of active users can be reduced by factors ranging from 4 to 600 by increasing indoor network coverage which, in turn, will be linked to increased exposure of non-users by a factor of 2 to 10. However, such an increase is by a factor 1000 lower than the typical exposure of active users. The results of this study are limited due to the network data that has been used and the definition of total exposure as stated earlier in this report. This study does not consider (i) the effect of upcoming massive MIMO and multi-user MIMO systems in 5G networks, (ii) alternative data transmission links - for instance the use of Wireless Local Area Network (WLAN) and (iii)millimeter wave frequencies in 5G mobile networks. In summary, the results of this study show that the user's own mobile device is the dominant source of exposure for the population of active mobile network users. Besides personal usage patterns, totl exposure is also closely linked to the network infrastructure. Generally speaking, a network that decreases the path loss by means of smaller cells and additional indoor coverage will help to reduce the total exposure of the population. bafu.admin.ch total-exposure-in-hypothetical-5g-mobile-networks-for-varied-topologies-and-user- scenarios.pdf.download.pdf/Modelling%20of%20Total%20Exposure%20in%20Hypothetical%205G%20Networks %20-%20Schlussbericht.pdf

Machine Enhanced Logic

AI evidence extraction

Structured fields below were generated from the paper’s abstract and metadata to make the evidence easier to inspect. Automated extraction may be incomplete or incorrect; verify important details against the original paper.

Study typeengineering

Effect directionmixed

PopulationGeneral population in modeled user scenarios (users and non-users)

Sample sizeNot supplied

ExposureRF · 3600 MHz · 6-minute average (psaSAR10g averaged over 6 minutes)

Evidence strengthLowConfidence: 78% · Peer reviewed: unknown

Main findings

Monte Carlo simulations (>200 scenarios) of hypothetical 5G networks found total exposure for active users is generally dominated by the person’s own mobile device; compared with non-users, modeled exposure increased by 6–10 dB for light users (100 MB uplink/day), 13–25 dB for moderate users (1 GB uplink/day), and 25–40 dB for heavy users. For non-users in urban areas, peak exposure was modeled to be driven more by nearby bystanders’ devices than by base stations (about 6 dB higher than from a nearby base-station antenna). A preliminary dosimetric analysis indicated tissue-averaged brain SAR at 3.6 GHz is reduced by >6× versus <1 GHz (and >2× versus 1.8–2 GHz) due to lower penetration depth, while this reduction may not apply to surface/near-surface tissues where peak SAR can be higher.

Outcomes measured

Total exposure defined as combined exposure from base stations, user's own device, and bystanders' devices
Local peak specific absorption rate spatially averaged over 10 g tissue (psaSAR10g), 6-minute average
Tissue-specific exposure as a function of frequency (e.g., brain vs surface tissues)
Effect of network topology (cell size, indoor coverage) and user scenarios (light/moderate/heavy use; non-users) on total exposure

Limitations

Hypothetical network scenarios; results based on modeling/Monte Carlo simulations rather than measured 5G deployments
Exposure metric defined as psaSAR10g averaged over 6 minutes; instant exposures not considered
Extrapolation used data measured in 4G systems and analysis of standards to estimate 5G exposures
Did not consider massive MIMO/multi-user MIMO systems
Did not consider alternative links such as WLAN
Did not consider 5G millimeter-wave frequencies

View raw extracted JSON

{
  "study_type": "engineering",
  "exposure": {
    "band": "RF",
    "source": "5G mobile network (hypothetical) including base stations, own device, and bystanders' devices",
    "frequency_mhz": 3600,
    "sar_wkg": null,
    "duration": "6-minute average (psaSAR10g averaged over 6 minutes)"
  },
  "population": "General population in modeled user scenarios (users and non-users)",
  "sample_size": null,
  "outcomes": [
    "Total exposure defined as combined exposure from base stations, user's own device, and bystanders' devices",
    "Local peak specific absorption rate spatially averaged over 10 g tissue (psaSAR10g), 6-minute average",
    "Tissue-specific exposure as a function of frequency (e.g., brain vs surface tissues)",
    "Effect of network topology (cell size, indoor coverage) and user scenarios (light/moderate/heavy use; non-users) on total exposure"
  ],
  "main_findings": "Monte Carlo simulations (>200 scenarios) of hypothetical 5G networks found total exposure for active users is generally dominated by the person’s own mobile device; compared with non-users, modeled exposure increased by 6–10 dB for light users (100 MB uplink/day), 13–25 dB for moderate users (1 GB uplink/day), and 25–40 dB for heavy users. For non-users in urban areas, peak exposure was modeled to be driven more by nearby bystanders’ devices than by base stations (about 6 dB higher than from a nearby base-station antenna). A preliminary dosimetric analysis indicated tissue-averaged brain SAR at 3.6 GHz is reduced by >6× versus <1 GHz (and >2× versus 1.8–2 GHz) due to lower penetration depth, while this reduction may not apply to surface/near-surface tissues where peak SAR can be higher.",
  "effect_direction": "mixed",
  "limitations": [
    "Hypothetical network scenarios; results based on modeling/Monte Carlo simulations rather than measured 5G deployments",
    "Exposure metric defined as psaSAR10g averaged over 6 minutes; instant exposures not considered",
    "Extrapolation used data measured in 4G systems and analysis of standards to estimate 5G exposures",
    "Did not consider massive MIMO/multi-user MIMO systems",
    "Did not consider alternative links such as WLAN",
    "Did not consider 5G millimeter-wave frequencies"
  ],
  "evidence_strength": "low",
  "confidence": 0.78,
  "peer_reviewed_likely": "unknown",
  "keywords": [
    "5G",
    "total exposure",
    "Monte Carlo simulation",
    "psaSAR10g",
    "specific absorption rate",
    "3.6 GHz",
    "base stations",
    "mobile device",
    "bystanders",
    "cell size",
    "indoor coverage",
    "Swiss FOEN",
    "IT'IS Foundation"
  ],
  "suggested_hubs": [
    {
      "slug": "5g-policy",
      "weight": 0.9,
      "reason": "Government-commissioned report modeling population exposure in hypothetical 5G networks and infrastructure choices."
    }
  ]
}

AI can be wrong. Always verify against the paper.

AI-extracted fields are generated from the abstract and metadata and may be incomplete or incorrect. This material is for informational purposes and is not medical advice.

Identifiers and source

DOI: Not supplied
PubMed: Not supplied

Read the original paper →

Related papers

← Back to the Research Library