Sustainability 14 00829

sustainability
Article
Measuring and Assessing Performance of Mobile Broadband
Networks and Future 5G Trends
Ayman A. El-Saleh 1 , Abdulraqeb Alhammadi 2, *, Ibraheem Shayea 3 , Nizar Alsharif 4 , Nouf M. Alzahrani 5 ,
Osamah Ibrahim Khalaf 6 and Theyazn H. H. Aldhyani 7
1 College of Engineering, A’Sharqiyah University (ASU), Ibra 400, Oman; ayman.elsaleh@asu.edu.om

2 Communication Systems and Networks Research Lab, Malaysia-Japan International Institute of Technology,
Universiti Teknologi Malaysia (UTM), Kuala Lumpur 54100, Malaysia
3 Department of Electronics and Communication Engineering, Faculty of Electrical and Electronics Engineering,
Istanbul Technical University (ITU), Istanbul 34467, Turkey; shayea@itu.edu.tr
4 Department of Computer Engineering and Science, Albaha University, Albaha 42331, Saudi Arabia;
Nizar@bu.edu.sa
5 College of Computer Science and Information Technology, Albaha University, Albaha 42331, Saudi Arabia;
Noufalzahrani@bu.edu.sa
6 Al-Nahrain Nanorenewable Energy Research Center, Al-Nahrain University, Baghdad 10071, Iraq;
usama.ibrahem@coie-nahrain.edu.iq
7 Applied College in Abqaiq, King Faisal University, P.O. Box 400, Al-Ahsa 31982, Saudi Arabia;
taldhyani@kfu.edu.sa
* Correspondence: abdulraqeb.alhammadi@utm.my
Abstract: Mobile broadband (MBB) is one of the critical goals in fifth-generation (5G) networks due
to rising data demand. MBB provides very high-speed internet access with seamless connections.
Existing MBB, including third-generation (3G) and fourth-generation (4G) networks, also requires

monitoring to ensure good network performance. Thus, performing analysis of existing MBB assists
Citation: El-Saleh, A.A.; Alhammadi,
mobile network operators (MNOs) in further improving their MBB networks’ capabilities to meet user
A.; Shayea, I.; Alsharif, N.; Alzahrani,
satisfaction. In this paper, we analyzed and evaluated the multidimensional performance of existing
N.M.; Khalaf, O.I.; Aldhyani, T.H.H.
Measuring and Assessing
MBB in Oman. Drive test measurements were carried out in four urban and suburban cities: Muscat,
Performance of Mobile Broadband Ibra, Sur and Bahla. This study aimed to analyze and understand the MBB performance, but it did
Networks and Future 5G Trends. not benchmark the performance of MNOs. The data measurements were collected through drive tests
Sustainability 2022, 14, 829. https:// from two MNOs supporting 3G and 4G technologies: Omantel and Ooredoo. Several performance
doi.org/10.3390/su14020829 metrics were measured during the drive tests, such as signal quality, throughput (downlink and
unlink), ping and handover. The measurement results demonstrate that 4G technologies were the
Academic Editor:
Manuel Fernandez-Veiga dominant networks in most of the tested cities during the drive test. The average downlink and
uplink data rates were 18 Mbps and 13 Mbps, respectively, whereas the average ping and pong loss
Received: 2 December 2021
were 53 ms and 0.9, respectively, for all MNOs.
Accepted: 6 January 2022
Published: 12 January 2022
Keywords: quality of experience; QoE; quality of service; QoS; mobile broadband; 5G; data rate
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations. 1. Introduction
Mobile broadband (MBB) networks are growing very fast, as they provide high-speed
internet access. The demand for data is rapidly increasing due to a huge number of
users accessing data through several cellular technologies and various types of internet
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
services. Spectrum demand for mobile communication systems is growing due to the
This article is an open access article
increase in the number of wireless technologies and applications. Intelligent technologies
distributed under the terms and are required to fully utilize and monitor current spectrums [1,2]. Dynamic spectrum access
conditions of the Creative Commons uses increase spectrum efficiency (SE) by adjusting radio resources. Performance and
Attribution (CC BY) license (https:// quality measurements become essential for all of the concerned mobile network operators
creativecommons.org/licenses/by/ to maintain and monitor the quality of service (QoS) of their existing networks. Therefore,
4.0/). fifth-generation (5G) networks provide enhanced MBB which supports high-speed data,
Sustainability 2022, 14, 829. https://doi.org/10.3390/su14020829 https://www.mdpi.com/journal/sustainability

Sustainability 2022, 14, 829 2 of 20
video streaming with low latency and seamless mobility. Real measurements of MBB
performance are affected by several factors, such as a variety of user devices, physical
impairments, mobility and accessibility configuration [3]. Therefore, mobile networks
need to be monitored to ensure user satisfaction in terms of QoS and quality of experience
(QoE). The international telecommunication union (ITU) introduced a new term called the
mean opinion score (MOS), which assesses mobile network providers to evaluate end-user
satisfaction by obtaining users’ opinions of a network’s performance [4]. The MOS metric
is widely used for several services and applications such as audio, audio–visual and video.
These services have various methods to assist in scoring their service quality.
Drive test is an important method for mobile operators to measure and assess the
capacity, coverage and QoS of mobile radio networks. The drive test consists of two ba-
sic types: user-equipment-based testing and benchmark testing. The former is based on
monitoring live end-user mobile devices such as smartphones and tablets, which provides
the mobile operators with the performance of their networks. The latter is used many user
devices and runs voice calls and data connections on other operators’ networks in any
particular area in order to determine the performance of each mobile network. Several
works have been carried out in the investigation and evaluation of the performance of the
existing MBB. In [5], the authors highlight the importance of QoE in cellular networks with
various radio access technologies (fourth-generation (4G), 5G and beyond), where they
provide the literature on the most advanced measurement methods in QoE. In addition,
the QoE is further investigated by different metrics and models for web QoE estimation [6].
The web QoE helps mobile network operators (MNOs) understand their customer’s usage
service patterns, perceive quality and point towards areas to improve. However, in other
studies [7,8], the authors investigate QoS and QoE by conducting an experimental study of
the current MBB supporting third-generation (3G) and 4G networks in Malaysia. A specific
application installed on a smartphone handset collected the drive test data from several
rural and urban regions. The measurement data of three MNOs are associated with several
performance indicators such as coverage, latency, satisfaction and speed for two MBB
services: web browsing and video streaming. The work in [9] analyzed data measurement
using several key performance indicators (KPIs) in 4G networks, such as signal quality and
download throughput. The drive test considered the actual road traffic conditions at a vehi-
cle speed of 30 km/h. The experimental results demonstrated that the achieved throughput
leads to different profiles in terms of time evolution. In [10], the authors investigated the
performance of nine MNOs in Europe during times when restrictions were in place due
to the COVID-19 outbreak. This investigation included several KPIs such as web QoE,
signal coverage, throughput and round-trip-time (RTT). The measurement results showed
approximate 46% increases in page load time at different times during the COVID-19
period. The findings indicated that the MNOs had managed their network performance
during the pandemic period, although some short-term performance degradations were
observable. In [11], the authors presented time and space mapping of outdoor electro-
magnetic field exposure induced by base station antennas in 4G cellular networks using
artificial neural networks. The data were obtained from electromagnetic field exposure
sensor networks. In [12], an embedded vehicle-to-everything platform was used to perform
drive tests on a public cellular network. The field measurements were conducted based on
existing passive network quality indicators and application-level information to forecast
uplink transmission power using a novel machine learning technique. Concering indoor
environments, the authors of [13] conducted MBB measurements to analyze the network
performance in indoor buildings of several areas in Malaysia. The measurement data were
collected from three MNOs using mobile smartphones for two types of MBB services: video
streaming and web browsing. The measurement results demonstrated that 80% of 4G and
3G networks coverage have good received signal strengths, whereas only 20% reach the
threshold level. In addition, 4G networks were the most accessed web and video cycles
compared with 3G networks. The authors of [14] conducted multiple measurements (static
and dynamic) in live LTE networks in terms of throughput to benchmark Austrian MNOs.
The finding showed that the static measurements are not applicable for benchmarking of
MNOs due to different small-scale fading patterns, whereas excellent benchmarking can be
obtained by using dynamic measurements.
In this paper, we investigate and evaluate the performance of existing MBBs (Omantel
and Ooredoo) in four cities in Oman. These cities are categorized as urban areas such as
Muscat and suburban such as Ibra, Bahla and Sur. Several KPIs were used to analyze the
MBBs, such as signal level and quality, throughput, ping rate and handover rate. This
investigation study provides valuable data measurements that can be used for future
network improvements and pinpointing during the deployment of 5G networks. The rest
of this paper is organized as follows: Section 2 provides an overall background on the MBB
networks and coverage. Section 3 describes the methodology and experimental design.
The performance of the MBBs is analyzed and evaluated in Section 4. Section 5 provides
study limitations and 5G trends. Section 6 concludes the paper.
2. Oman Mobile Broadband

The demand for data is rapidly increasing due to a huge number of users accessing
data through several cellular technologies and various types of internet services. Spectrum
demand for mobile communication systems is growing due to the increase in wireless
technologies and applications. Ericsson reported that the gulf cooperation council countries
are among the most advanced information and communication technology (ICT) markets
in the world [15]. The statistics demonstrated that over 90% of mobile subscriptions were
for MBB at the end of 2020. It is estimated that this percentage will increase up to 95% in
2026. The report showed that 4G networks are the dominant technology, making up about
80% of subscriptions at the end of 2020.
In the Oman telecommunications regulatory authority (TRA) report, the Oman’s tele-
com industry saw significant repercussions due to COVID-19, as evidenced by increasing
demand for services during lockdown and reduced average expenditure owing to various
special offers by service providers [16]. The report discussed the MBB usage in gigabytes
(GBs): there was a 27.61% growth in the total GBs consumed, although active MBB sub-
scriptions dropped by 8.72%. However, in the second quarter of 2020, 59.8% of the land
area was covered by at least a 3G mobile network, whereas 99.3% of the population was
covered with at least a 3G mobile network [16].
Figure 1 demonstrates the coverage network map of 2G, 3G, 4G and 5G mobile
networks for Omantel and Ooredoo in Muscat, the capital city of Oman. These data are
based on the nPerf application, which collects the data from tests carried out by users’
devices [17]. The application retains tests with a maximum geolocation precision of 50 m for
coverage data. From Figure 1, it can be seen that 4G networks are the dominant technology
for all MNOs. It displays the collected data for two years between November 2019 and
November 2021, in which approximately 136,045 and 140,024 pieces of data were collected
for Omantel and Ooredoo, respectively. The data were obtained from tests performed
by nPerf app users. These tests were carried out in the field under real-world settings.
Omantel recorded a few 5G coverage points (violet color), whereas there is no record for
5G coverage in Ooredoo. To provide complete 5G network coverage, MNOs would need to
build more base stations, especially in populated areas.
For wider coverage and to fulfil the demand for MBB services, MNOs need to deploy
a large number of base stations. Thus, MNOs should build new base stations with various
cellular technologies based on the type of planning area to serve a large population of users
with a good QoS level.
Sustainability 2022,
Sustainability 2022, 14,
14, 829
x FOR PEER REVIEW 44 of 23
of 20
(a)
(b)
Figure 1.
Figure 1. MBB
MBB coverage
coverage in
in Muscat
Muscat city
city [17]:
[17]: (a)
(a) Omantel,
Omantel, (b)
(b) Ooredoo.
Ooredoo.
For wider
Table 1 andcoverage
Figure and to fulfil
2 show the the demand
growth fornumber
in the MBB services,
of sitesMNOs need to deploy
per technology and
a large number
average SE of theofmain
base stations.
MNOs inThus,OmanMNOsfor 2015should build
to 2020. Thenew base stations
statistical withobtained
data were various
cellular
from technologies
prominent based
agencies, on as
such thethetype of planning
operators’ annualarea to serve
reports, a large MTC
regulators, population of
[18] and
users with a good QoS level.
Ooredoo [19]. The existing 2G, 3G and 4G networks have several versions: GSM, GPRS and
EDGE,Table 1 andHSUPA,
HSDPA, Figure HSPA
2 show andtheHSPA+,
growthand in the
LTEnumber of sites
and LTE-A, per technology
respectively. and
The typical
average SE
average SE of
forthe
eachmain MNOs inisOman
technology for 20154.87
0.17 bps/Hz, to 2020. The statistical
bps/Hz and 23.16 data werefor
bps/Hz obtained
2G, 3G
from4G
and prominent
networks, agencies, such as
respectively. In the
thisoperators’
regard, the annual
averagereports,
SE forregulators,
all sites is MTC [18] and
calculated by
Ooredoo [19].
multiplying theThe existing
number 2G,overall
of the 3G and 4Gwith
sites networks haveaverage
the typical several SEversions:
for eachGSM, GPRS
technology.
and total
The EDGE, SE HSDPA, HSUPA,
and SE growth areHSPA
combinedand HSPA+,
to presentandtheLTE and LTE-A,
expected resultsrespectively.
for Oman. The
typical average
Figure SE for
3 shows theeach technology
SE growth is 0.17
of the two bps/Hz,
considered4.87MNOs
bps/HzinandOman.23.16Thebps/Hz for
existing
2G, 3G were
MNOs and 4G networks,
labeled respectively.
as X and In this regard,
Y due to confidential the average
issues; this study SEdoes
for all
notsites is calcu-
benchmark
lated
the by multiplying
MNOs’ performance. the number
The data of the overall
indicate sites
that thewith the typical
SE and growthaverage
ratio willSEradically
for each
technology.
increase Thethe
within total
nextSEyears.
and SE Thegrowth
averagearespectrum
combinedefficiency
to presentperthe expected
site and growthresults for
ratios
Oman. operators in 2020 was 10.06 bps/Hz and 1.89, respectively. It can be seen that this
overall
increase is due to the outcome of the continuous migration and upgrading of the mobile
cellular systems from old to new technologies.
Sustainability 2022, 14, x FOR PEER REVIEW 5 of 23
Table 1. Growth in number of sites per technology and ASE.
Table 1. Growth in number of sites per technology and ASE. SE Growth Ratio
Sites Number Per Technology Average SE
Year Overall
Sites Number Per Technology (bps/Hz)
Year 2G 3G 4G Average SE Technology
SE Growth Ratio
2G (bps/Hz) Overall Technology
2015 4083 3745 3G 1285 4G 5.33
2016 2015 4279 4083 4403 3745 20081285 6.41
5.33 1.20
2016 4279
2017 4396 4805 4403 30862008 6.41
7.78 1.20
1.46
2017 4396 4805 3086 7.78 1.46
2018 2018 4426 4426 5226 5226 38813881 8.58
8.58 1.61
1.61
2019 2019 4498 4498 5399 5399 42704270 8.89
8.89 1.67
1.67
2020 2020 4609 4609 5923 5923 58265826 10.06
10.06 1.89
1.89
7000 2G 3G 4G
6000
5000
Sites Number
4000
3000
2000
1000
0
2015 2016 2017 2018 2019 2020
Year
Figure 2. Sites number growth for all network technologies.
Figure 2. Sites number growth for all network technologies.

12
Figure 3 shows theOperator
SE growth
X of theOperator
two considered
Y MNOs in Oman. The existing
MNOs were
10 labeled as X and Y due to confidential issues; this study does not benchmark
the MNOs’ performance. The data indicate that the SE and growth ratio will radically
Spectrum Effciency [bps/Hz]
increase 8within the next years. The average spectrum efficiency per site and growth ratios
overall operators in 2020 was 10.06 bps/Hz and 1.89, respectively. It can be seen that this
increase is due to the outcome of the continuous migration and upgrading of the mobile
6
cellular systems from old to new technologies.
0
2015 2016 2017 2018 2019 2020
Year
Figure 3. SE growth for each MNO.
Figure 3. SEEach
growth for each MNO. service provider has a different strategy for future mobile
operator/mobile
network deployment and development. This plan is usually confidential and cannot be
Each operator/mobile
publicly serviceany
shared. Therefore, provider
relatedhas a different
data cannot bestrategy for futureAcquire
easily obtained. mobile net-
real data
workfrom
deployment and development. This plan is usually confidential and
operator/mobile service providers regarding network performance and cannot be pub-
strategic
licly shared. Therefore,
plans for the futureany related data
deployment of cannot
mobile be easily obtained.
networks Acquire real data from
is challenging.
operator/mobile service providers regarding network performance and strategic plans for
the future deployment of mobile networks is challenging.
3. Methodology and Experimental Design

This section presents the test methodology of the measurement campaign to evaluate
3. Methodology and Experimental Design

This section presents the test methodology of the measurement campaign to evaluate
the existing MNOs with several performance metrics relevant to the user experience. The
data collection was carried out using a commercial android application developed by
Gyokov Solutions called “G-NetTrack”, installed in two Samsung Galaxy handsets [20].
In addition, several applications were tested in terms of performance metrics, recording
logfile, continuity of testing and stability. This application supports drive tests for outdoor
scenarios and provides a wide range of features such as map visualization, cell scanning
loading cells, cell measurement for serving and neighbor cells. G-NetTrack has been used
in numerous research papers such as [21–23], which means it is considered a reliable
application for collecting data measurements. Figure 4 visualizes the information of data
measurement for cell and network information.
Figure4.4.Example
Figure Exampleofofmeasurement
measurementinformation
informationby
byG-NetTrack.
G-NetTrack.
Table22displays
Table displaysthe
thecollected
collecteddata
dataduring
duringthethedrive
drivetest
testthat
thatwas
wassaved
savedin
inthe
thelog
logfile
file
aftercompleting
after completingthethemeasurements.
measurements.There Therewere
weresix sixmain
mainKPIs
KPIsused
usedtoto analyze
analyze the
the MNOs’
MNOs’
performance,namely:
performance, namely:signal
signallevel
level(RSRP),
(RSRP),signal
signal quality
quality (RSRQ),
(RSRQ), channel
channel quality
quality indica-
indicator
tor (CQI),
(CQI), ping,
ping, throughput
throughput (downlink
(downlink (DL) (DL)
andand uplink
uplink (UL)
(UL) data)
data) andand handoverrate.
handover rate.All
All
theseKPIs
these KPIswere
wererecorded
recordedininthe
thelog
logfile
fileaccording
accordingtotoMNOs.
MNOs.
Table 3 shows the setting parameters of the data sequence for each mobile device. Both
Table 2.were
devices Termsset
and
updescription of thesettings
with the same recordedinlog file. to achieve a fair comparison. The data
order
sequence started with the ping test first and then was followed by a DL/UL test, where
Term Description
there was a 1 s time pause between test cycles. A larger size of the download file was
Longitude
selected Current location
which could not be downloaded fullylongitude in decimal
in the allocated format.time in order to
download
Latitude
achieve a higher download speed. Current location latitude in decimal format.
Speed Vehicle speed in km/h.
Type Type of network technology: 2G/3G/4G.
Signal strength in dBm recorded as: RXLEV for 2G, RSCP for 3G and RSRP
LEVEL
for 4G. For simplicity, RSRP was used as a signal level for all networks.
Signal quality of the network in dB and recorded as: RXQUAL for 2G, ECNO
QUAL for 3G and RSRQ for 4G. For simplicity, RSRQ was used as signal quality for
all networks.
CQI Channel quality indicator measure over 4G network only.
Table 2. Terms and description of the recorded log file.
Term Description
Longitude Current location longitude in decimal format.
Latitude Current location latitude in decimal format.
Speed Vehicle speed in km/h.
Type Type of network technology: 2G/3G/4G.
Signal strength in dBm recorded as: RXLEV for 2G, RSCP for 3G and RSRP
LEVEL
for 4G. For simplicity, RSRP was used as a signal level for all networks.
Signal quality of the network in dB and recorded as: RXQUAL for 2G,
QUAL ECNO for 3G and RSRQ for 4G. For simplicity, RSRQ was used as signal
quality for all networks.
CQI Channel quality indicator measure over 4G network only.
Measured the required time to send an amount of data and receive a
Ping
response in ms.
DL/UL Downlink/uplink data transfer speed in kbps.
The process of transferring voice calls or data sessions from one serving
Handover
cell to a target cell.
Table 3. Setting parameters of data sequence.
Parameter Setting
Ping URL www.google.com (accessed on 28 August 2021).
Ping number 10
Ping packet size 56 bytes
Ping interval 1000 ms
Ping sequence time 10 s
http://ipv4.download.thinkbroadband.com (accessed on
Upload URL
28 August 2021).
Upload time 10 s
http://ipv4.download.thinkbroadband.com/1GB.zip (accessed
Download URL
on 28 August 2021).
Download file size 1 GB
Download time 10 s
Pause between test 1s
Multithread No
Simultaneous UL/DL Yes
The data measurements were performed in four cities: Muscat, Ibra, Bahla and Sur.
In this regard, the MBB performance of the two national MNOs (Omantel and Ooredoo)
was investigated and evaluated. However, this study aimed to analyze and understand
the MBB performance, but it did not benchmark the performance of MNOs. Thus, the
existing MNOs were labeled as X and Y in the discussed and demonstrated results. Prepaid
subscriber identification module (SIM) cards with the same data package were used for
each MNO to ensure peer to peer comparison. Figure 5 displays the general methodology
of the data measurements and analysis. The two mobile devices were fixed on a mobile
phone stand holder inside a car, as shown in Figure 6. Then, the mobile devices collected
the data from the MNO towers and then stored them on the device’s memory that could be
used for analysis after completing the data campaign. Figure 7 displays the experimental
testbed (measurement area) where the data measurements were taken on the orange route.
Sustainability
Sustainability2022,
2022, 14,
14, x829
FOR PEER REVIEW 9 8ofof23
20
Figure
Figure5.5.AAgeneral
generalmethodology
methodologyof
ofdata
datacollection
collectionand
andanalysis.
analysis.
Figure 5. A general methodology of data collection and analysis.
Figure 6. Data measurement during drive test.

Figure6.6.Data
Figure Datameasurement
measurementduring
duringdrive
drivetest.
test.
Figure 8 shows the block diagram of the general data measurement. In the drive
test, first, the app was set up, setting several parameters of data sequences such as ping
upload time and URL, data rate time (down/upload) and tested file with one gigabyte in
size. Once these parameters were set, the drive test logged the data measurement from the
starting point and stopped at the ending points. The signal level and quality were measured
continuously during the drive test. In addition, several events such as handover and cell
reselection were recorded during the data measurements. All these data measurements
were associated with the timestamp and GPS coordinates (longitude and latitude).
Sustainability
2022, 14, 14,
829x FOR PEER REVIEW 10 of 923
of 20
(a) (b)
(a) (b)
(c) (d)
Figure 7. Tracking route of measurement area: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
Figure 8 shows the block diagram of the general data measurement. In the drive test,
first, the app was set up, setting several parameters of data sequences such as ping upload
time and URL, data rate time (down/upload) and tested file with one gigabyte in size.
Once these parameters were set, the drive test logged the data measurement from the
starting point and stopped at the ending points. The signal level and quality were meas-
(c) (d)
ured continuously during the drive test. In addition, several events such as handover and
cell reselection
Figure
Figure 7. Trackingwere
7. Tracking routerecorded duringarea:
of measurement
route of measurement the(a)
area: data
(a) measurements.
Muscat,
Muscat, All
(b) Ibra, (c) Sur
(b) Ibra, (c) Sur these
and
and (d) data
Bahla.measure-
(d) Bahla.
ments were associated with the timestamp and GPS coordinates (longitude and latitude).
Figure 8 shows the block diagram of the general data measurement. In the drive test,
first, the app was set up, setting several parameters of data sequences such as ping upload
time and URL, data rate time (down/upload) and tested file with one gigabyte in size.
Once these parameters were set, the drive test logged the data measurement from the
starting point and stopped at the ending points. The signal level and quality were meas-
ured continuously during the drive test. In addition, several events such as handover and
cell reselection were recorded during the data measurements. All these data measure-
ments were associated with the timestamp and GPS coordinates (longitude and latitude).
Figure 8. Block diagram of the general measurement methodology.

Figure 8. Block diagram of the general measurement methodology.
4. Performance Evolution
The evolution of existing MBBs is essential to ensure that a network provides a high-
quality user experience. This experimental research was conducted to analyze the actual
performance of the current national MNOs in Oman. It is an extension of the work that
was carried out in our previous work in [24]. All the measurements were conducted during
Figure
the 8. Block
daytime diagram
with of theoperating
the same general measurement
system and methodology.
test sequences for both operators in each
city in order to achieve a fair comparison. The measurement results are demonstrated and
discussed for each KPI: RSRP, RSSI, RSRQ, CQI, ping, throughput and handover rate. The
measurements were collected in the daytime for all MNOs. The car speed was limited up to
The evolution of existing MBBs is essential to ensure that a network provides a high-
quality user experience. This experimental research was conducted to analyze the actual
performance of the current national MNOs in Oman. It is an extension of the work that
was carried out in our previous work in [24]. All the measurements were conducted dur-
Sustainability 2022, 14, 829 ing the daytime with the same operating system and test sequences for both operators 10in
of 20
each city in order to achieve a fair comparison. The measurement results are demonstrated
and discussed for each KPI: RSRP, RSSI, RSRQ, CQI, ping, throughput and handover rate.
The measurements were collected in the daytime for all MNOs. The car speed was limited
70 km/h for all outdoor scenarios. Figure 9 illustrates the car speeds during the drive test.
up to 70 km/h for all outdoor scenarios. Figure 9 illustrates the car speeds during the drive
Ittest.
can It
becan
observed that the speed was maintained below 70 km/h and sometimes reached
be observed that the speed was maintained below 70 km/h and sometimes
0reached
km/h at0 traffic lights.
km/h at traffic lights.
80
70
60
Speed (Km/h)
50
40
30
20
10
0
1
159
317
475
633
791
949
1107
1265
1423
1581
1739
1897
2055
2213
2371
2529
2687
2845
3003
3161
3319
3477
3635
3793
3951
4109
4267
4425
4583
4741
4899
5057
5215
5373
5531
5689
5847
Time (s)
Figure 9. Car speed versus time.

Figure 9. Car speed versus time.
Figure10
Figure 10visualizes
visualizesthe
theevent
event recordings
recordings and
and measured
measured KPIs
KPIs (RSRP, RSRQ and
(RSRP, RSRQ and CQI)
CQI) of
Sustainability 2022, 14, x FOR PEERof one
REVIEW
one operator.
operator. TheThe route
route colors
colors represent
represent the the measured values of each KPI. In
12 the
23 fol-
measured values of each KPI. In theoffollowing
lowing subsections,
subsections, the performance
the performance for eachfor
MNOeachisMNO is analyzed
analyzed and discussed
and discussed according
according to KPIs.
to KPIs.
(a) (b)
(c) (d)
Figure 10. Tracking route of measurement area: (a) event recordings, (b) RSRP, (c) RSRQ and (d)
Figure 10. Tracking route of measurement area: (a) event recordings, (b) RSRP, (c) RSRQ and (d) RSSI.
RSSI.
4.1. Signal Level and Quality

Figure 11 displays the measured data of RSRP and RSSI for the two MNOs. Both of
the operators’ average RSRP and RSSI levels are between −85 dBm and −57 dBm and −81
4.1. Signal Level and Quality

Figure 11 displays the measured data of RSRP and RSSI for the two MNOs. Both of
the operators’ average RSRP and RSSI levels are between −85 dBm and −57 dBm and
−81 dBm and −68 dBm, respectively. This finding indicates that both operators have good
network coverage in the measurement area. However, the minimum RSRP for all tested
cities and operators is below −120 dBm. This value represents the mobile device at the
cell edge where RSRP levels for usable signal typically range from −75 dBm (near 13
Sustainability 2022, 14, x FOR PEER REVIEW
to of
base
23
station) to −120 dBm (at the cell edge of network coverage). The data measurements show
that all the MNOs cover all the tested cities with 4G networks.
Muscat -150 Ibra

-150
Operator X Operator Y Operator X Operator Y
Signal level (dBm)

Signal level (dBm)
-100 -100
-50 -50
0 0
Min Average Max Min Average Max Min Average Max Min Average Max
RSRP RSSI RSRP RSSI
(a) (b)
-150 Sur -150

Bahla
Signal level (dBm)
Signal level (dBm)
-100 -100
-50 -50
0 0
RSRP RSSI RSRP RSSI
(c) (d)
Figure
Figure 11.
11. Measured
Measured RSRP
RSRP and
and RSSI
RSSI for
for the
the two
two MNOs:
MNOs: (a)
(a) Muscat,
Muscat, (b)
(b) Ibra,
Ibra, (c)
(c) Sur
Sur and
and (d)
(d) Bahla.
Bahla.
Figure
Figure 12
12 shows
shows the
the levels
levels of
of RSRQ
RSRQ inin the
the measured
measured areas
areas for
for the
the two
two MNOs.
MNOs. Many
Many
resource
resource blocks
blocks with
with measured
measured RSRP
RSRP and
and RSSI
RSSI were
were used
used over
over the
the same
same bandwidth
bandwidth for
for
each
each MNO.
MNO.The TheRSRQ
RSRQvalue
valuehelps thethe
helps base station
base perform
station cell reselections
perform or intra-inter
cell reselections or intra-
handovers. The ﬁgure
inter handovers. shows shows
The figure the average RSRQ level
the average RSRQ that varied
level that between −9 dB and
varied between −9−11
dB
dB,
andwhere
−11 dB,thewhere
RSRQthe
achieved
RSRQ aachieved
maximum and minimum
a maximum value of −5value
and minimum dB and −20
of − dB,and
5 dB re-
spectively.
−20 dB, respectively.
Figure 13 illustrates the level of CQI for the measured MNOs in the four cities. The
MNOs used CQI to monitor the channel quality between evolved NodeBs (eNB) and mobile
devices. The CQI reported value ranges from 0 to 15, indicating the modulation and coding
level at which a mobile device could operate on. The CQI provides information about the
quality conditions of the communication channel. It can be seen that the average CQI levels
vary between 11 and 8 for all MNOs in all cities, where all MNOs obtain a maximum value
of CQI of approximately 15. The high CQI level indicates a high RSRQ. A more reliable
connection is usually the result of higher RSRQ and CQI. Thus, the CQI is a key metric for
LTE systems, where MNOs typically use it to determine the relationship between radio link
conditions and throughput.
Sustainability2022,
Sustainability 2022,14,
14,829
x FOR PEER REVIEW 14
12 of 23
of 20
-25 -25
Muscat Ibra
-20 -20
RSRQ Level (dB)
RSRQ Level (dB)

-15 -15
-10 -10
-5 -5
0 0
Min Average Max Min Average Max
(a) (b)
-25 -25
Sur Bahla
-20 -20
RSRQ Level (dB)
RSRQ Level (dB)
-15 -15
-10 -10
-5 -5
0 0
Min Average Max Min Average Max
(c) (d)
Figure 12. Average rate of RSRQ level for the measured MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d)
Figure 12. Average rate of RSRQ level for the measured MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
Bahla.
4.2. Data Rate
Figure
Data 13 is
rate illustrates
the mostthe level of CQI
important for the measured
monitoring metric in MNOs
mobile in the four
cellular cities. The
networks; it
MNOs used CQI to monitor the channel quality between evolved NodeBs
is used to evaluate internet speed performance. It defines the amount of data delivered (eNB) and mo-
bile devices.
over a network Theduring
CQI reported value
a specified rangestypically
period, from 0 tomeasured
15, indicating theper
in bits modulation and
second (bps).
coding level at which a mobile device could operate on. The CQI
Figure 14 displays the throughput of the measured MNOs in terms of minimum, maximum provides information
about
and the quality
average for allconditions
cities. Theoffigure
the communication
demonstrates the channel.
DL/UL It can be seen that
throughput for the
eachaverage
MNO
in the four measured cities. The maximum data rates in DL are approximately morea
CQI levels vary between 11 and 8 for all MNOs in all cities, where all MNOs obtain
maximum
than 100 Mbpsvalue inof CQI
Ibra and of Sur
approximately
and below 50 15.Mbps
The high CQI level
in Muscat andindicates
Bahla foraboth
highMNOs.
RSRQ.
However, all measured cities achieve approximate similar maximum UL data ratesCQI
A more reliable connection is usually the result of higher RSRQ and CQI. Thus, the for
is a key
both metric
MNOs. Infor LTE systems,
addition, both MNOswhereachieve
MNOs typically use it tothroughput
a good average determine the relationship
in Ibra and Sur
between radio
compared withlink conditions
Muscat and The
and Bahla. throughput.
reason for this result is that Muscat and Bahla are
very crowded cities compared with Ibra and Sur, which require a bandwidth boost.
4.3. Latency (Ping, Packet Loss)

Ping measures the reaction time of speed connection and is recorded as ping count,
where a lower ping rate is better than a higher rate. In practice, the RTT of 4G networks
tends to be in the range of 30 to 100 ms. Figure 15 shows the ping of the measured MNOs
in terms of average, standard deviation and packet loss. The results show that the MNOs
achieve an average ping rate lower than 50 ms for all cities within the accepted range,
indicating good speed connections. Overall, operator Y achieves a lower average ping rate
compared with operator X in the measured cities. Besides, it also obtains an average packet
loss of approximately zero compared with operator X in all cities except Muscat.
20 20
Muscat Ibra
Operator X Operator Y
15 Operator X Operator Y
15
CQI
10
CQI
10
5 5
0 0
Min Average Max StDv
Min Average Max StDv
(a) (b)
20 20
Sur Bahla
15 15
CQI
CQI
10 10
5 5
0 0
Min Average Max StDv Min Average Max StDv
(c) (d)
Figure 13. Average rate of CQI level for the measured MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d)
Figure 13. Average rate of CQI level for the measured MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
Bahla.
4.2. Data Rate

120
Muscat Data rate is the most important monitoring Ibra
50 metric in mobile cellular networks; it is
used to evaluate internet speed performance.
100 It defines the amount of data delivered over
40
Throghput (Mbps)
a networkOperator
during X a specified period, typically measured in bits perOperator
secondX (bps). Figure
Throghput (Mbps)
80
30 14 displays the throughput of the measured MNOs in terms of minimum, Operatormaximum
Y and
Operator Y 60
average for all cities. The figure demonstrates the DL/UL throughput for each MNO in the
20 four measured cities. The maximum data 40 rates in DL are approximately more than 100
10
Mbps in Ibra and Sur and below 50 Mbps in Muscat and Bahla for both MNOs. However,
20
all measured cities achieve approximate similar maximum UL data rates for both MNOs.
0 In addition, both MNOs achieve a good0 average throughput in Ibra and Sur compared
Min Average Max Min Average
with Muscat Max The reason for this
and Bahla. Min Average
result MaxMuscat
is that Min and
Average Maxare very
Bahla
DL crowded citiesULcompared with Ibra and Sur, which require
DL a bandwidthULboost.
(a) (b)
Sur Bahla
120 35
100 30
Operator X Operator X
Throghput (Mbps)
Throghput (Mbps)
25
80 Operator Y
Operator Y 20
60
15
40
10
20 5
0 0
DL UL DL UL
(c) (d)
Figure 14. Average data rate (DL and UL) in the MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
Figure 14. Average data rate (DL and UL) in the MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
4.3. Latency (Ping, Packet Loss)
Ping measures the reaction time of speed connection and is recorded as ping count,
where a lower ping rate is better than a higher rate. In practice, the RTT of 4G networks
tends to be in the range of 30 to 100 ms. Figure 15 shows the ping of the measured MNOs
in terms of average, standard deviation and packet loss. The results show that the MNOs
14, 829
14 of 23
of 20
Muscat 60
100 Ibra
50
80
Operator X
Operator X 40
Ping (ms)
60 Operator Y
Ping (ms)
Operator Y 30
40
20
20 10
0 0
Average StDv Loss Average StDv Loss
(a) (b)
120 300
Sur Bahla
100 250
Operator X Operator X
80 200
Operator Y
Ping (ms)
Ping (ms) Operator Y

60 150
40 100
20 50
0 0
Average StDv Loss Average StDv Loss
(c) (d)
Figure 15.
Figure 15. Average
Average ping
pingrate
rateand
andpacket
packetloss
lossininthe
theMNOs:
MNOs:(a)
(a)Muscat,
Muscat,(b)
(b)Ibra,
Ibra,(c)(c)Sur
Sur and
and (d)(d)
Bahla.
Bahla.
4.4. Handover
4.4. Handover
In mobile cellular networks, handover is a process of transferring voice calls or data
In mobile
sessions from onecellular networks,
serving cell to handover is a Several
a target cell. process mobility
of transferring
issuesvoice
occurcalls or data
during the
sessions from
handover one such
process, serving cell to a target
as ping-pong handovercell. and
Several
radiomobility issues
link failure. occur
These during
issues the
signifi-
handover
cantly process,
affect such as ping-pong
the communication quality handover
in termsand radio
of long link failure.time
interruption These
andissues signif-
throughput
icantly affect [25].
degradation the communication
There are two types qualityofinhandovers,
terms of long interruption
namely time andhandover
intra-frequency through-
put degradation
and inter-frequency [25].handover.
There are Thetwo former
types ofoccurs
handovers,
among namely intra-frequency
base stations handover
that operate in the
and inter-frequency
same network technology.handover.The The
latterformer
occursoccurs
among among base stations
base stations that operate
that operate in the
in different
same network
network technology.
technologies. FigureThe 16latter occursthe
illustrates among base number
handover stations of
that
theoperate in different
two MNOs for all
network
cities. Thetechnologies. Figure 16
data measurements illustrates
show that 4Gthe handover
networks number
were of the two
the dominant MNOs
serving for all
network
cities. The data measurements show that 4G networks were the dominant serving network
in most cities except Bahla city, where operator X has experience with 3G networks. From
in most cities except Bahla city, where operator X has experience with 3G networks. From
Figure 16, the most recorded handovers occur within the same technology (4G-to-4G han-
Figure 16, the most recorded handovers occur within the same technology (4G-to-4G
dover) except Bahla city (operator X), where 4G-to-3G handovers are detected due to radio
handover)
link except Bahla
communication city (operator
exchanged from 4GX), to where 4G-to-3GGenerally,
3G technology. handovers are detected
operator due to
Y achieves a
radio link
lower communication
handover number than exchanged
operatorfromX in 4G to 3Gexcept
all cities technology.
Bahla. Generally,
A possibleoperator
reason for Y
achieves
this findinga lower
is thehandover
operators’number
settingsthan operatormanagement,
of mobility X in all citieswhich
exceptdepend
Bahla. Aonpossible
several
reason for this finding is the operators’ settings of mobility management, which depend
factors such as time-to-trigger, handover margin, handover events and handover offset.
on several factors such as time-to-trigger, handover margin, handover events and hando-
4.5. Analysis Summary
ver offset.
Several performance metrics were analyzed for two MNOs in four cities in Oman.
Figure 17 exhibits a comparison among cities taking account of signal level and quality,
throughput, ping rate and handover number. Figure 17a shows that MNOs in all cities have
almost the same signal levels and qualities. Higher UL and DL throughput is achieved by
operator X and Y, respectively, in Ibra city, as shown in Figure 17b. In Figure 17c, operator X
in Bahla obtains a higher ping rate compared with other cities. Additionally, operator X in
Muscat achieves a higher handover number, as shown in Figure 17d. Table 4 summarizes
the signal level and quality (RSRP, RSRQ and CQI). Table 5 summarizes throughput, ping
rate and handover number. Digits that are in bold type refer to best performance. All the
MNOs cover the measured area with 4G at measurement time with good signal quality
and data rates. These demonstrated results of each operator depend on several factors such
as bandwidth capacity, the number of sites, the number of active users and the type of
subscribers. These factors significantly affect the data rate and QoS that may improve or
degrade the network performance.
350 Muscat Ibra

140
300 120
Number of Handover
Number of Handover
250 100
200 80
150 60
100 40
50 20
0 0
(a) (b)
Sur 120
Bahla
250
100
200
Number of Handover
Number of Handover
80 4G_4G
150 3G_3G
60 4G_3G
100
40
50
20
0
Operator X Operator Y 0
(c) (d)
Figure 16. Handover number in the MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
Figure 16. Handover number in the MNOs: (a) Muscat, (b) Ibra, (c) Sur and (d) Bahla.
4.5. Analysis Summary

Table 4. Summary of the average signal qualities.
Several performance metrics were analyzed for two MNOs in four cities in Oman.
Figure
RSRP (dBm)17 exhibits a comparisonRSRQamong
(dB) cities taking account of signal
CQI level and quality,
City MNO
Min throughput,
Avg. ping
Max rate Minand handover
Avg number.Max Figure 17a shows
Min Avgthat MNOsMax in allSdv.
cities
have almost the same signal levels and qualities. Higher UL and DL throughput is
X −112 −85.98 −54 −20 −9.27 −3 1 8.48 15 2.55
Muscat achieved by operator X and Y, respectively, in Ibra city, as shown in Figure 17b. In Figure
Y −117 −80.63 −51 −20 −10.03 −5 1 10.66 15 3.51
17c, operator X in Bahla obtains a higher ping rate compared with other cities. Addition-
−114 −84.90
Ibra
X
ally, operator−X60in Muscat
−18 −9.62
achieves −5
a higher handover4 number, as shown15in Figure
10.80 2.34
17d.
Y −119 −82.70 −52 −20 −10.27 −5 1 8.62 15 3.20
Table 4 summarizes the signal level and quality (RSRP, RSRQ and CQI). Table 5 summa-
X −113 −85.70
rizes −44 ping rate
throughput, −20 and handover
−11.11 −5
number. Digits
1 that are in bold type
8.77 15 refer 2.59
to best
Sur
Y −117 −78.66 −44 −20 −10.85 −4 1 7.00 14 2.63
performance. All the MNOs cover the measured area with 4G at measurement time with
Bahla
X −117 good signal quality
−57.66 −24 and−20
data rates. These demonstrated
−10.96 −5 1 results of each operator
10.69 15 depend
2.30
Y −112 − 81.46 − 58 − 20 − 10.15 − 4 3 10.90 15
on several factors such as bandwidth capacity, the number of sites, the number of active 2.98
users and the type of subscribers. These factors significantly affect the data rate and QoS
that may improve or degrade the network performance.
14, 829
16 of 23
of 20
(a) (b)
(c) (d)
Figure 17.
Figure 17. Comparison
Comparisonamong
amongmeasured
measuredareas with
areas respect
with to (a)
respect signal
to (a) level
signal andand
level quality, (b)
quality,
throughput, (c) ping and (d) handover number.
(b) throughput, (c) ping and (d) handover number.
Table 4. Summary of the average signal qualities.

Table 5. Summary of average throughput, ping rate and handover number.
RSRP (dBm) RSRQ (dB) CQI
CityThroughput
MNO (Mbps)
Min Avg. Max Min Avg Max Min AvgHandover
Ping (ms) Max Sdv.No.
City MNO
Avg DL Avg ULX −112DL −85.98
Max UL Max−54 −20
Avg −9.27 Stdev−3 1
Loss 8.48 15
Total 2.55
Muscat
X 0.93 12.76 Y −117
8.36 −80.63
18.89−51 −20
46.94 −10.03
94.55 −5 1
1.12 10.66 15
309 3.51
Muscat
Y 7.48 15.14 X −114
45.84 −84.90
20.93−60 −18
35.86 −9.6244.66 −5 4
0.18 10.80 15
189 2.34
Ibra
X 9.26 15.54 Y −119
104.01 −82.70
22.77−52 −20
48.18 −10.27
27.30 −5 1
2.18 8.62 15
129 3.20
Ibra
Y 20.90 13.80 X −113
107.54 −85.70
21.53−44 −20
41.71 −11.11
41.53 −5 1
0.05 8.77 1588 2.59
Sur 13.67
Sur
X 7.75 Y −117
112.17 −78.66
31.77−44 −20
49.70 −10.85
17.80 −4 1
1.76 7.00 14
234 2.63
Y 11.53 13.71 100.68 20.77 38.15 105.16 0 175
X −117 −57.66 −24 −20 −10.96 −5 1 10.69 15 2.30
X 3.77 Bahla 7.46 Y −112 −81.46 −58 −20 −10.15
29.21 21.82 128.87 238.22 −4 3
2.14 10.90 1536 2.98
Bahla
Y 6.23 15.04 27.81 23.41 39.38 27.00 0 111
5. Study limitations and 5G Trends

5.1. Study Limitations
This investigation study presented the performance of existing MBB networks. Several
limitations can be discussed as follows:
1. Limited drive test: The data measurements were collected in four cities, whereas larger
areas would be better but more time consuming. Thus, this work can be extended to
other areas, especially densely populated cities. Moreover, this work can also include
rural areas which are outside the densely populated urban areas in towns or cities.
2. Indoor scenarios: This study can be further extended to include the performance of
MBB in indoor environments such as shopping malls while considering single and
multi-floor scenarios. Indoor coverage would become much more important due
to the high demands of modern usage cases. Thus, further investigation in indoor
environments will provide an overview of how MBB performs in a complicated
internal building’s structure that can interfere with radio frequencies.
3. One-time drive test: This study was carried out during the daytime where the cell
traffic load was at a normal capacity. Hence, drive tests during peak and peak-off
hours can be included in further investigations. In the peak period, most of the
subscribers are likely to be online and demand more resources, whereas the data rate
demand is low in the off-peak period.
4. MBB services: Due to application limitations, this study was limited to two types of
MBB services: web browsing and file (DL and UL) tests. Other services tests such
as video streaming and voice can be included in future study, but it will require an
application that can support these services.
5. Beyond MBB services: Several factors other than MBB services have not been consid-
ered in this study, such as tariffs, prices, data packages, policies, privacy, billing, etc.
These factors are not important in the analysis of network performance, but it will be
more useful for the benchmarking stages.
6. Measurement time: The measurements of this study were collected once for each area
where the network performance of MNOs was not measured with various climatic
conditions with multiple drive tests. Thus, a one-time drive test may not provide the
actual network performance of each MNOs. Therefore, this study can be extended
further to multiple, longer drive tests for each area.
7. Auto technology: In this study, the mobile device chose the available network technol-
ogy (i.e., 2G, 3G or 4G) to connect to a serving cell. However, this did not provide the
network performance of each technology of MNO independently. It can be recom-
mended to include a locked technology scenario where the mobile device is locked to
one technology at each measurement time.
8. User Mobility: The data measurements were collected with the medium mobility of
car speed (≤70 kh/h). Therefore, various mobility scenarios could be considered for
the drive test, such as low, medium and high speeds. These scenarios lead to the
provision of more details on the network performance with respect to user mobility.
9. 5G networks: This study focused on existing MBB networks (3G and 4G) because
there are only limited commercial 5G MBB networks deployed in Oman. Therefore,
the current study can be extended to involve 5G MBB using supported mobile phones
and applications.
5.2. 5G Trends
The Oman TRA allocates the frequency band 3400–3600 MHz to the licensed MNOs
to be used for 5G networks with a bandwidth of 100 MHz for each MNO. Omantel and
Ooredoo are allocated frequency bands of 3400–3500 MHz and 3500–3600 MHz, respectively.
Recently, both operators have deployed their 5G networks at the non-standalone stage.
Approximately 1300 5G sites have been deployed in 11 governates for the two MNOs, as
shown in Figure 18.
5.2.5. G Trends
The Oman TRA allocates the frequency band 3400–3600 MHz to the licensed MNOs
to be used for 5G networks with a bandwidth of 100 MHz for each MNO. Omantel and
Ooredoo are allocated frequency bands of 3400–3500 MHz and 3500–3600 MHz, respec-
Sustainability 2022, 14, 829
tively. Recently, both operators have deployed their 5G networks at the non-standalone
18 of 20
stage. Approximately 1300 5G sites have been deployed in 11 governates for the two
MNOs, as shown in Figure 18.
300
250
200
Number of 5G Sites
150
100
50
0
Muscat Dhofar Musandam Al Buraymi Ad Dakhiliyah Al Batinah NorthAl Batinah South Ash Sharqiyah Ash Sharqiyah Ad Dhahirah Al Wusta
South North
Governorates of Oman
Figure 18. 5G sites distributions among Oman governorates.

Figure 18. 5G sites distributions among Oman governorates.
In addition,
In addition, they
they launched
launched super-fast
super-fast and
and reliable
reliable 4G
4G and
and 5G
5G fixed
fixed wireless
wireless (FWA)
(FWA)
access. A trial demonstrated Omantel and Ericsson’s ability to deliver
access. A trial demonstrated Omantel and Ericsson’s ability to deliver multi-gigabit multi-gigabit
speeds
speeds
of of the
the FWA FWAa using
using carrierabandwidth
carrier bandwidth of 800
of 800 MHz at MHz at 26
26 GHz. GHz. Ooredoo
Ooredoo andinitially
and Nokia Nokia
initially covered 3000 homes in city centers with the 5G FWA services using
covered 3000 homes in city centers with the 5G FWA services using Nokia equipment [11]. Nokia equip-
ment
In [11]. InTRA
addition, addition,
Oman TRA Oman
declared thatdeclared that a third
a third operator operator
mobile licensemobile license
(Vodafone) (Voda-
would be
fone)
in would soon
operation be in to
operation
supportsoon to support
and improve andperformance.
MBB improve MBB performance.
6. Conclusions
6. Conclusions
This study
This study presented
presented aa performance
performance analysis
analysis ofof the
the existing
existing national
national MNOs
MNOs in in Oman.
Oman.
A drive test was conducted in urban and suburban areas. The network performance of the
MNOs was wasanalyzed
analyzedbyby several
several performance
performance metrics
metrics suchsuch as signal
as signal levelquality
level and and quality
(RSRP,
(RSRP, RSRQ and CQI), DL/UL throughput, ping rate and handover number. The meas-
RSRQ and CQI), DL/UL throughput, ping rate and handover number. The measurement
urement
results resultsthat
showed showed that theinMNOs
the MNOs in have
all cities all cities havethe
almost almost
samethe same
signal signal
and andlevels.
quality qual-
ity levels. Higher UL and DL throughput is achieved in Ibra city, whereas operator
Higher UL and DL throughput is achieved in Ibra city, whereas operator X obtains X ob-
a higher
tains rate
ping a higher ping rate
than other citiesthan other Additionally,
in Bahla. cities in Bahla. Additionally,
operator operator
X achieves X achieves
a higher handover a
number in Muscat city. In addition, the 4G networks were the dominant networks during
the drive test except for operator X in Bahla city; the 3G network was recorded on one long
route. Overall, MNOs achieve an average DL and UL throughput of approximately 8 Mbps
and 13 Mbps, respectively, with an average ping rate and loss of 53 ms and 0.9, respectively,
in the considered cities. It can be concluded that all the MNOs performed well and
maintained good coverage and capacity during the drive test. In future work, this study
can be further extended to include the performance of MBB in rural areas and indoor
environments such as shopping malls. More importantly, the performance of 5G MBB
networks will be considered in our future research.
Author Contributions: Conceptualization, A.A.E.-S. and I.S.; Data curation, A.A.E.-S.; Formal analy-
sis, A.A.; Funding acquisition, N.A., N.M.A. and O.I.K.; Investigation, A.A.; Methodology, A.A. and
I.S.; Supervision, A.A.E.-S.; Writing—review & editing, A.A., A.A.E.-S. and T.H.H.A. All authors
have read and agreed to the published version of the manuscript.
Funding: The research leading to these results has received funding from the Research Council (TRC)
of the Sultanate of Oman under the Block Funding Program with agreement no. TRC/BFP/ASU/01/2019.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement: Not applicable.
Acknowledgments: This research study was sponsored by the Universiti Teknologi Malaysia through
the Professional Development Research University Grant (No. 05E92).
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Alhammadi, A.; Roslee, M.; Alias, M.Y. Fuzzy logic based negotiation approach for spectrum handoff in cognitive radio network.
In Proceedings of the 2016 IEEE 3rd International Symposium on Telecommunication Technologies (ISTT), Kuala Lumpur,
Malaysia, 28–30 November 2016.
2. Roslee, M.; Alhammadi, A.; Alias, M.Y.; Anuar, K.; Nmenme, P. Efficient handoff spectrum scheme using fuzzy decision making
in cognitive radio system. In Proceedings of the 2017 3rd International Conference on Frontiers of Signal Processing (ICFSP),
Paris, France, 6–8 September 2017.
3. Lobo, B.J.; Alam, M.R.; Whitacre, B.E. Broadband speed and unemployment rates: Data and measurement issues. Telecommun.
Policy 2020, 44, 101829. [CrossRef]
4. ITU. Series P: Telephone Transmission Quality, Telephone Installations, Local Line Networks, Vocabulary for Performance, Quality of Service
and Quality of Experience; T P.10/G.100; International Telecommunication Union Telecommunication Standardization Sector:
Geneva, Switzerland, 2017.
5. Bouraqia, K.; Sabir, E.; Sadik, M.; Ladid, L. Quality of experience for streaming services: Measurements, challenges and insights.
IEEE Access 2020, 8, 13341–13361. [CrossRef]
6. Jahromi, H.Z.; Delaney, D.T.; Hines, A. Beyond first impressions: Estimating quality of experience for interactive web applications.
IEEE Access 2020, 8, 47741–47755. [CrossRef]
7. Shayea, I.; Azmi, M.H.; Ergen, M.; El-Saleh, A.A.; Han, C.T.; Arsad, A.; Rahman, T.A.; Alhammadi, A.; Daradkeh, Y.I.; Nandi, D.
Performance analysis of mobile broadband networks with 5g trends and beyond: Urban areas scope in malaysia. IEEE Access
2021, 9, 90767–90794. [CrossRef]
8. Shayea, I.; Ergen, M.; Azmi, M.H.; Nandi, D.; El-Salah, A.A.; Zahedi, A. Performance analysis of mobile broadband networks
with 5G trends and beyond: Rural areas scope in Malaysia. IEEE Access 2020, 8, 65211–65229. [CrossRef]
9. Imoize, A.L.; Orolu, K.; Atayero, A.A.-A. Analysis of key performance indicators of a 4G LTE network based on experimental
data obtained from a densely populated smart city. Data Brief 2020, 29, 105304. [CrossRef] [PubMed]
10. Rajiullah, M.; Khatouni, A.S.; Midoglu, C.; Alay, Ö.; Brunstrom, A.; Griwodz, C. Mobile network performance during the
COVID-19 outbreak from a testbed perspective. In Proceedings of the 14th International Workshop on Wireless Network Testbeds,
Experimental evaluation & Characterization, London, UK, 25 September 2020.
11. Wang, S.; Wiart, J. Sensor-Aided EMF Exposure Assessments in an Urban Environment Using Artificial Neural Networks. Int. J.
Environ. Res. Public Health 2020, 17, 3052. [CrossRef]
12. Falkenberg, R.; Sliwa, B.; Piatkowski, N.; Wietfeld, C. Machine learning based uplink transmission power prediction for LTE
and upcoming 5G networks using passive downlink indicators. In Proceedings of the 2018 IEEE 88th Vehicular Technology
Conference (VTC-Fall), Chicago, IL, USA, 27–30 August 2018.
13. Shayea, I.; Rahman, T.A.; Azmi, M.H.; Han, C.T.; Arsad, A. Indoor network signal coverage of mobile telecommunication
networks in West Malaysia: Selangor and Johor Bahru. In Proceedings of the 2017 IEEE 13th Malaysia International Conference
on Communications (MICC), Johor Bahru, Malaysia, 28–30 November 2017.
14. Raida, V.; Svoboda, P.; Rupp, M. On the Inappropriateness of Static Measurements for Benchmarking in Wireless Networks. In
Proceedings of the 2020 IEEE 91st Vehicular Technology Conference (VTC2020-Spring), Antwerp, Belgium, 25–28 May 2020.
15. Ericsson. Ericsson Mobility Report; Ericsson: Stockholm, Sweden, 2021.
16. TRA. Telecom Sector Indicators Report; Telecommunications Regulatory Authority: Muscat, Oman, 2020.
17. Perf, 3G/4G/5G Coverage Map, Oman. 2021. Available online: https://www.nperf.com/en/map/OM/-/223624.Omantel-
Mobile/signal/ (accessed on 25 November 2021).
18. MTC. Annual Report 2019; Ministry of Technology and Commnications: Muscat, Oman, 2019; pp. 1–62.
19. Ooredoo. Capital Markets Day, Ooredoo Oman Updates. 2019. Available online: https://www.ooredoo.com/wp-content/
uploads/2015/10/Capital-Markets-Day-Ooredoo-Oman-CEO-Ian-Dench-.pdf (accessed on 25 July 2021).
20. Solutions, G. G-NetTrack Pro. Available online: https://gyokovsolutions.com/g-nettrack/ (accessed on 28 August 2021).
21. Elsherbiny, H.; Nagib, A.M.; Abou-zeid, H.; Abbas, H.M.; Hassanein, H.S.; Noureldin, A.; Sediq, A.B.; Boudreau, G. 4G LTE
Network Data Collection and Analysis along Public Transportation Routes. In Proceedings of the GLOBECOM 2020-2020 IEEE
Global Communications Conference, Taipei, Taiwan, 7–11 December 2020.
22. De Andrade, H.G.V.; Ferreira, C.C.; da Silva Filho, A.G. Latency analysis in real lte networks for vehicular applications. In
Proceedings of the 2016 VI Brazilian Symposium on Computing Systems Engineering (SBESC), João Pessoa, Paraíba, Brazil, 1–4
November 2016.
23. Raca, D.; Quinlan, J.J.; Zahran, A.H.; Sreenan, C.J. Beyond throughput: A 4G LTE dataset with channel and context metrics. In
Proceedings of the 9th ACM Multimedia Systems Conference, Amsterdam, The Netherlands, 12–15 June 2018.
24. Jahdhami, M.A.A.; El-Saleh, A.; Alhammadi, A.; Shayea, I. Performance Analysis of Mobile Broadband Networks in Ibra City,
Oman. In Proceedings of the IEEE International Conference on Artificial Intelligence and Big Data Analytics (ICAIBDA 2021),
Jawa Barat, Indonesia, 27–29 October 2021.
25. Abdulraqeb, A.; Mardeni, R.; Yusoff, A.; Ibraheem, S.; Saddam, A. Self-optimization of handover control parameters for mobility
management in 4G/5G heterogeneous networks. Autom. Control Comput. Sci. 2019, 53, 441–451. [CrossRef]

Sustainability 14 00829

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sustainability 14 00829

Uploaded by

Copyright:

Available Formats

sustainability

1 College of Engineering, A’Sharqiyah University (ASU), Ibra 400, Oman; ayman.elsaleh@asu.edu.om

Sustainability 2022, 14, 829. https://doi.org/10.3390/su14020829 https://www.mdpi.com/journal/sustainability

2. Oman Mobile Broadband

Sustainability 2022, 14, 829 5 of 20

Table 1. Growth in number of sites per technology and ASE.

Figure 2. Sites number growth for all network technologies.

Figure 2. Sites number growth for all network technologies.

Figure 3. SE growth for each MNO.

3. Methodology and Experimental Design

3. Methodology and Experimental Design

Table 2. Terms and description of the recorded log file.

Table 3. Setting parameters of data sequence.

Figure 6. Data measurement during drive test.

Figure 8. Block diagram of the general measurement methodology.

Figure 9. Car speed versus time.

4.1. Signal Level and Quality

4.1. Signal Level and Quality

Muscat -150 Ibra

Signal level (dBm)

RSRP RSSI RSRP RSSI

-150 Sur -150

RSRQ Level (dB)

4.3. Latency (Ping, Packet Loss)

4.2. Data Rate

Ping (ms) Operator Y

350 Muscat Ibra

4.5. Analysis Summary

Table 4. Summary of the average signal qualities.

5. Study limitations and 5G Trends

Figure 18. 5G sites distributions among Oman governorates.

Informed Consent Statement: Not applicable.

You might also like