Multi-objective optimisation of human exposure for various 5G network topologies in Switzerland

2022 · Mixed · Low evidence

← Research stream

MixedLow evidencePaper 619

Study snapshot

Publicationmanual · 2022

Study typeengineering

Reported directionMixed

Evidence ratingLow

Abstract

Multi-objective optimisation of human exposure for various 5G network topologies in Switzerland Castellanos G, De Gheselle S, Martens L, Kuster N, Joseph W, Deruyck M, Kuehn S. Multi-objective optimisation of human exposure for various 5G network topologies in Switzerland. Computer Networks.216, 2022. doi: 10.1016/j.comnet.2022.109255. Abstract The constant increase in the required user capacity and the evolution of wireless network technologies impact the exposure that users experience from wireless networks. This paper evaluates various 5G network topologies regarding human exposure, mobile communication quality, and sustainability. We assess human exposure, based on a novel Exposure Ratio (ER) metric, in 5G networks that include Massive Multiple-Input Multiple- Output (MaMIMO) and compare them with existing 4G deployments in three environments in Switzerland. The quality and sustainability of mobile communication are evaluated by extrapolating data rates from mobile operators to the year 2030. A multi-objective optimisation algorithm is implemented to design the 5G network topologies, maximising the user coverage while minimising the downlink (DL) and uplink (UL) exposure. An extensive set of simulations investigated three municipalities, three operators plus one unified network, three use cases (UL/DL data rates), three scenarios (indoor and outdoor coverage), and two optimisation methods. The study results confirm that the human exposure in a 5G network is dominated by the UL being ten times larger than the DL exposure. Furthermore, comparing a 5G deployment with 10 times the traffic capacity of a real 4G network, DL exposure increases by 36% on average, and UL exposure decreases by up to 75% depending on the scenario. Regarding indoor coverage versus outdoor only, our results show that DL exposure can be reduced by a factor of 10 if only outdoor coverage is targeted. Finally, the study concludes that from the human exposure perspective, the ideal network should use 5G MaMIMO and be optimised for both UL and DL exposure. sciencedirect.com

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

PopulationNot supplied

Sample sizeNot supplied

ExposureRF

Evidence strengthLowConfidence: 74% · Peer reviewed: yes

Main findings

Simulation-based multi-objective optimisation of 5G network topologies in three Swiss environments found that uplink exposure dominated downlink exposure by about a factor of 10. Compared with a 4G network, a 5G deployment with 10× traffic capacity showed downlink exposure increased by 36% on average while uplink exposure decreased by up to 75% depending on scenario; targeting outdoor-only coverage could reduce downlink exposure by a factor of 10 versus indoor coverage.

Outcomes measured

Human exposure (Exposure Ratio metric; downlink and uplink exposure)
Mobile communication quality (data rates extrapolated to 2030)
Sustainability (as evaluated in the optimisation framework)
User coverage

Limitations

Simulation study; real-world measurements not described in abstract
Frequencies, power levels, and exposure limits/benchmarks not specified in abstract
Exposure metric (ER) is novel; external validation not described in abstract
Results depend on modeled scenarios (municipalities, operators, use cases, indoor/outdoor assumptions)

View raw extracted JSON

{
  "study_type": "engineering",
  "exposure": {
    "band": "RF",
    "source": "5G network (including Massive MIMO) compared with 4G deployments",
    "frequency_mhz": null,
    "sar_wkg": null,
    "duration": null
  },
  "population": null,
  "sample_size": null,
  "outcomes": [
    "Human exposure (Exposure Ratio metric; downlink and uplink exposure)",
    "Mobile communication quality (data rates extrapolated to 2030)",
    "Sustainability (as evaluated in the optimisation framework)",
    "User coverage"
  ],
  "main_findings": "Simulation-based multi-objective optimisation of 5G network topologies in three Swiss environments found that uplink exposure dominated downlink exposure by about a factor of 10. Compared with a 4G network, a 5G deployment with 10× traffic capacity showed downlink exposure increased by 36% on average while uplink exposure decreased by up to 75% depending on scenario; targeting outdoor-only coverage could reduce downlink exposure by a factor of 10 versus indoor coverage.",
  "effect_direction": "mixed",
  "limitations": [
    "Simulation study; real-world measurements not described in abstract",
    "Frequencies, power levels, and exposure limits/benchmarks not specified in abstract",
    "Exposure metric (ER) is novel; external validation not described in abstract",
    "Results depend on modeled scenarios (municipalities, operators, use cases, indoor/outdoor assumptions)"
  ],
  "evidence_strength": "low",
  "confidence": 0.74,
  "peer_reviewed_likely": "yes",
  "keywords": [
    "5G",
    "4G",
    "Massive MIMO",
    "MaMIMO",
    "Exposure Ratio (ER)",
    "uplink exposure",
    "downlink exposure",
    "multi-objective optimisation",
    "network topology",
    "Switzerland",
    "simulations",
    "coverage",
    "data rates"
  ],
  "suggested_hubs": [
    {
      "slug": "5g-policy",
      "weight": 0.6,
      "reason": "Evaluates human exposure implications of 5G network topologies (including Massive MIMO) versus 4G."
    }
  ]
}

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: 10.1016/j.comnet.2022.109255
PubMed: Not supplied

Read the original paper →

Related papers

← Back to the Research Library