-
Comparison of the average global
exposure of a population induced by a
macro 3G network in urban, suburban
and rural areas
Y. Huang(1,2,3), A. Krayni(1,2), A. Hadjem(1,2), J. Wiart (1,2), C. Person(2,3) and N. Varsier(1,2)
(1) Orange, Issy les Moulineaux, France
(2) Whist Lab, common laboratory between Institut Mines-Telecom and Orange, France
(3) Institut Mines-Telecom Telecom Bretagne, Brest, France
Speaker : Yuanyuan HUANG
-
recherche & développement
Plan
Study purpose
European FP7 Project LEXNET :Exposure Index (EI)
Analysis of mobile usage
Network traffic planning tool
Average global exposure of a population
Conclusion
Perspectives
-
recherche & développement
What is the purpose of the study
Moving from individual to population
exposure
Moving from maximum assessment to
real life quantification
Moving from one link exposure to
both downlink and uplink exposure
Base stations
Wireless handheld devices
Uplink
exposure
Downlink
exposure
1 URSI AT-RASC 2015
-
recherche & développement
European FP7 Project LEXNET : Low EMF Exposure Future Network
9 countries 17 partners
LEXNET Objectives: Minimizing the Human
population EMF exposure
• Defining of a global exposure metric named
Exposure Index (EI) covers the day-to-day
exposure of a population in a given area incurred
by entire wireless network from base stations to
individual devices
• Investigating and prosing architecture and
effective mechanisms reduce EMF human
exposure without compromising quality of
services
Duration: November 2013 ‒ October 2015
2 URSI AT-RASC 2015
-
recherche & développement
The Exposure Index (EI), a new RF exposure metric
The EI chain of exposure
3 URSI AT-RASC 2015
-
recherche & développement
Mathematical expression of EI
Life Segmentation Data
Up-to-date life segmentation surveys collected in the different participated countries were used
4
ICT Usage Data are obtained through measurements of Key
Performance Indicators (KPIs) by sensors inside
the network
Reference SAR Data provided by numerical 3D
electromagnetic SAR simulation softwares using anatomical
human body models.
Tx and Rx Network traffic planning
tools: ASSET and Odyssee tools
URSI AT-RASC 2015
-
recherche & développement
Traffic measurement probe
Radio Network controller
Structured Query Language
France 3G network
Key Performance Indicator
Analysis of mobile usage: data from 3G Orange Network
• Covered area
-- urban, suburban, rural
• Types of data
-- voice, data traffic, Sms, Mms
• Registration duration
-- one week in Nov. 2013
5 URSI AT-RASC 2015
-
recherche & développement
Segmentation of mobile usage: user profiles
Heavy users:
top users consuming 80% of the
total data (% of total data >> % of
users)
Moderate users:
users consuming 15% of the total
data (% of total data ≈ % of users)
Light users:
users consuming 5% of the total
data (% of total data
-
recherche & développement
Segmentation of mobile usage: summary of user profiles
Orange France 3G network
Rural Suburban Urban Aggregation
tcom voldata [MB]
tcom voldata [MB]
tcom voldata [MB]
tcom voldata [MB]
Heavy
users
33min 103 31min 71 27min 68 29min 73
Moderate
users
4min 9 6min 8 4min 5 5min 7
Light
users
42s 0,5 71s 0,8 41s 0,3 53s 0,5
Rural: Same average communication time compared to
other types of areas but highest volume data exchanges!
7 URSI AT-RASC 2015
-
recherche & développement
Network traffic simulations: Input/Output
Network
traffic
planning
tool
Output: Average UL power and average DL power density in the considered
geographical areas:
• Tx Power for a phone call usage
• Tx Power for a data usage
• Rx Power density
𝑃 𝑇𝑋_indoor_voice_day
𝑃 𝑇𝑋_indoor_voice_night
𝑃 𝑇𝑋_outdoor_voice_day
𝑃 𝑇𝑋_outdoor_voice_night
𝑃 𝑇𝑋_indoor_data_day
𝑃 𝑇𝑋_indoor_data_night
𝑃 𝑇𝑋_outdoor_data_day
𝑃 𝑇𝑋_outdoor_data_night
𝑆 RXinc_indoor_day
𝑆 RXinc_indoor_night
𝑆 RXinc_outdoor_day
𝑆 RXinc_outdoor_night
Input
Population density
Repartition of the profile users
(heavy%, moderate%and light%)
Traffic usage (tcom&voldata) during
the day and night for each profile
user
Macro 3G Network in the
considered area
8 URSI AT-RASC 2015
-
recherche & développement
Network traffic planning tools
A multi-systems simulator for radio network can evaluate:
• Quality of radio network: ex. coverage
• QoS: ex. good communication probability
• Tx³ et Rx⁴ powers
Simulator operating on real data:
• geographical area
• network environment
• density of traffic
• propagation models
High Resolution map data (colors gradient base on height)
Traffic raster data with resolution 25 meters
Propagation
3G network environment
[3] Tx power, power emitted by the mobile phone
[4] Rx power, power received by the mobile phone from base stations 9 URSI AT-RASC 2015
-
recherche & développement
Macro 3G Orange network in urban suburban and rural areas
Urban area: Lyon central city
Population Density :10369 habs/km
53 sites corresponding to 154 cells in
the study area of 14 km²
Population category Percentage
(France)
Children (-15 y.o) 13,9%
Young people (15 – 29
y.o)
32,8%
Adults (30 – 59 y.o) 38,2%
Seniors (+60 y.o) 15,1%
10
Suburban area: Cergy city
Population Density :5404 habs/km²
34 sites corresponding to 90 cells in
the study area of 90 km²
Population category Percentage
(France)
Children (-15 y.o) 22,2%
Young people (15 –
29 y.o)
31,4%
Adults (30 – 59 y.o) 39,1%
Seniors (+60 y.o) 7,3%
Rural area: rural area near Metz
Population Density :2851 habs/km²
25 sites coresponding to 62 cells in
the study area of 94 km²
Population category Percentage
(France)
Children (-15 y.o) 16,1%
Young people (15 – 29 y.o) 28,1%
Adults (30 – 59 y.o) 38,8%
Seniors (+60 y.o) 17%
URSI AT-RASC 2015
-
recherche & développement
EI evaluation: suburban macro 3G scenarios
11
Indoor Day
time Rx
Tx data
Tx voice
URSI AT-RASC 2015
Cergy city
Life segmentation and ICT usage
• 2 time periods: day 8am – 6 pm / night 6pm – 8am
• 4 categories of population: children, young people,
adults, seniors
• Surveys: time spend indoor/outdoor per population
category
• Surveys: proportions of users/non-users of mobile
phones per population category
• Hypothesis: Proportions of heavy, moderate and light
users per population category
• Usage: 70% indoor/ 30% outdoor; 40% of the 3G
connections; 35% as the Orange subscribers percent
of total market
• Heavy, moderate and light usage of mobile phones
Tx/Rx powers
SAR reference values
Whole-body SAR induced by mobile phones and by macro
base stations for adult and child for different postures
(standing and sitting) and 2 different usages (voice and
data).
-
recherche & développement
time period
Life segmentation data
Percentages of
usages
Mobile usages
EI integration on the urban, suburban and rural 3G macro scenarios
EMF exposure values
Average global exposure
of a population
15
Tx/Rx
EIWBSAR
[W/kg]
France %UL
exposure
Urban 3,36e-7 92%
Suburban 2,01e-7 89%
Rural 1,90e-7 98%
12
Exposure Index Equation
URSI AT-RASC 2015
-
recherche & développement
Conclusion
User profiles depend on the considered geographical area, small
differences exist among them
We evaluated the average exposure of a population to EMF induced
by a 3G macro network in three different geographical areas (dense
urban, suburban and rural), with respect to different user profiles
EI calculated for the population in urban area is the highest
Uplink exposure has important weight to the global exposure for a
3G network
Downlink exposure in urban area is important because of the high
base stations density
URSI AT-RASC 2015 13
-
recherche & développement
Perspectives
Many data collected from other countries reflecting ICT usages of
the population and of technologies
The aim of evaluate EI is to use it as a KPI for network optimization
in terms of both downlink and uplink exposures to EMF.
We can compare cities, populations and technologies by evaluating
EI
-- collaborative assessment with Lexnet partners
14 URSI AT-RASC 2015
-
Thank you for your attention