Int. j. inf. tecnol.

https://doi.org/10.1007/s41870-020-00455-3

ORIGINAL RESEARCH

Analysis of audio and video quality of voice over LTE (VoLTE)

call
Smita Avinash Lonkar1 • KT V. Reddy2

Received: 5 June 2019 / Accepted: 6 April 2020

Ó Bharati Vidyapeeth’s Institute of Computer Applications and Management 2020

Abstract LTE is a broadband wireless communication and Uplink. Such knowledge will help software manufac-
standard, including a voice service that is upgraded to 5 G, turers on the IMS stack to enhance video quality and
with the core network technology. In recent years Voice identify audio output systems on MOS basis. In this,
over Internet Protocol (VoIP) became more accessible to research discuss VoLTE and Video over LTE (ViLTE)
customers because of the increasing competence of Internet performance issues. In the ITRI VoLTE test environment,
services and the development of voice call applications. we analyse VoLTE voice quality in measurement experi-
According to its increased capacity, data rates, and lower ments. We will also pose ViLTE Call with a check in a
latency, LTE meets the demand for high data rates for FET industrial mobile network in multimode or multi-band
many mobile users. Voice over LTE (VoLTE) is the Cel- interworking. Our study provides information on the
lular network system for transmitting voice over TCP. To quality of the VoLTE and ViLTE services for operators.
network operators or system manufacturers, the quality of
service (QoS) assessment of VoLTE coverage is essential. Keywords Voice over LTE (VoLTE)  Quality of service
This article presents the evaluation analysis of video (QoS)  Key performance indicators (KPIs)  Mean opinion
quality QoS Key Performance Induction (KPI) and the sore (MOS)
POLQA Mean Audio Performance Rating in the VoLTE
application. This measurement can be done for various
video quality analysis display formats and packeting mode. 1 Introduction
Audio quality analysis in the VoLTE call is done by
measuring the statistical and distribution information for In the IP network’s situation, voice communication is
Uplink and Downlink Mean opinion score (MOS). MOS transmitted by IP packets through a data connection. The
for video quality assessment (Average) is conducted. methods used for this application primarily to web-based
Graphical analyses of the video rating distributions verify data transmission, use IP packets, which also carry voice
the calculated QoS KPIs for video calls. The calculation traffic. Achieving the required QoS is of the utmost
analysis finishes with the verification of the option in importance for the accessibility of an adequate bandwidth
VoLTE Call for video codec screen formats and packeting to send and receive network and speech packets. Following
mode. Audio performance can be defined as the audio 2010, VoIP applications like Skype, What Sapp and G talk
quality of the test device by the visual display of Downlink will use the internet on 2 G or 3 G networks. The appli-
cation path of the OTT program is not discriminated from
& Smita Avinash Lonkar
other IP data traffic. Thus QoS can be severely affected by
[email protected] voice interaction. [1] 3GPP mainly worked on achieving a
KT V. Reddy
high data performance with a decreased latency through
[email protected] LTE network creation. LTE is only an IP network, and the
ability to move conventional voice is not relevant. A sep-
1
U.M.I.T., S.N.D.T. Women’s University, Mumbai, India arate approach was needed for LTE networks for the
2
PRES, Loni, India transport of switched circuit calling. This approach is

123
 Int. j. inf. tecnol.

typically known as ‘‘VoLTE’’ to transmit speech over IP on QoS analysis of KPI vendors and device manufacturers
LTE networks. VoLTE converts speech into a data stream is essential information. The main purpose of this paper is
that is sent via the data link. For user experience, it is to evaluate different screen formats and packet mode
crucial to provide voice services connected to the data metrics for improving video quality and to examine sta-
channel for other services such as video streaming, web tistical data for improving audio quality. Such data will
browsing, and the public media. Therefore, as VoIP is the help suppliers of video quality control tools in the IMS
best service, the fundamental difference between VoLTE stack and product identification dependent on MOS
and VoIP is; VoLTE has the potential to validate QoS from efficiency.
the beginning. [2] The successful implementation of the IP The technological aspects of VoLTE implementations
Multimedia System (IMS) framework in the LTE network on the existing circuit-changed (CS) and packet-changed
is a sure scheme of end to end QoS. The IMS supports (PS) networks were discussed in this article. The speech
various access and multimedia services and has evolved as function of the VoLTE system was implemented via the
a regular packet core (EPC) architecture [3–5]. IMS platform in the LTE network. A better description of
3GPP committed with GSMA IR.92 IMS Profile [6] and the carrying voice on the LTE network makes IMS derived
GSMA IR.94 IMS Profile [7] when it provides high-quality speech usually available.
IMS-based telephony products via the LTE mobile access This paper also provides output analysis based on video
network. This defines the optimal selection of the current evaluation information and VoLTE call quality. The KPIs
3GPP technologies to provide network infrastructure are assessed for QCIF, QVGA and VGA displaying for-
manufacturers, service providers, and smartphone devel- mats and FU and SNU packeting modes. The video quality
opers with optimized Cellular mobile and multimedia test cases help us analyze the effects on the key KPIs for
solutions. different parameters. Session Score is calculated as the
The use of VoLTE has presented mobile providers with arithmetical mean in all parts of a session. The sum is
several benefits because speech is the primary source of determined. In all situations with the right frame rate under
revenue. In comparison to current CS networks, VoLTE varying RF settings, the weighted session rating and the
improves bandwidth efficiency and reduces network costs. video quality are determined. Image rating distribution is
Thus, VoLTE is an effective option for voice in 4 G net- graphically evaluated for all configurations and video
works. [8, 9]. quality test cases checked. The checks were conducted for
VoLTE increases end-user understanding by superior VoLTE call audio quality in many situations. This paper
quality of experience (QoE). In contrast, the VoLTE plat- explains the sample two cases.
form only supports multiple speech codecs frequencies As far as potential work is concerned, this paper pro-
AMR with both wireless AMR-WB, AMR-NB, and poses to test audio video sync (AV-sync) KPIs in all for-
enhanced voice services (EVS). In turn, it provides an extra mats, and to analyze the impact on VoLTE call quality.
wide-band AMR-WB range. The paper is organized appropriately. Section 3 in short
To give up to 20 kHz audio range, the 3GPP Rel-12 explains different mechanisms to bring voice traffic to a
implemented EVS codec. [10] Super wide-band EVS LTE network. Section 2 reviews existing research articles
(SWB) with 13,2kbps offers an equal bit rate for voice focused on VoLTE QoS. Call procedure in VoLTE is
quality to AMR and AMR-WB with hard chips and packet briefed in Sect. 4. Section 5 describes the main video and
failures. audio call performance indicators (KPIs). The findings
VoLTE provides high definition speech (HD) commu- were presented and addressed in Sect. 6 and the assump-
nications end-to-end (E2E) services to provide a better user tions are eventually clarified in Sect. 7.
experience. A collection of QoS category IDs (QCIs) has
been established by 3GPP and are used by entities to
provide the minimum QoS specifications for their end user 2 Literature survey
system forbearers of particular preferences and attributes.
For other products, distinct QoS rates are offered, which In this chapter, researchers on VoLTE and QoS discuss
are assured by VoLTE. To build an E2E signal and a some essential approaches to science. A short overview is
carrier stream, E2E QoS needs the help of cell terminals, given of some improvements to the current methods.
communication networks, and core networks. This can Yunhan Jack Jia et al. [1] introduced the commercially
improve the QoS commitment of voice services and implemented VoLTE’s quality description and contrasted it
enhance the experience of customers. A practical VoLTE to the standard 3 G call, Over the Top (OTT) VoIP call.
QoS analysis is evaluated in MOS to determine voice VoLTE excels in the performance of clips but lags behind
quality. [11]. conventional 3 G communication efficiency. Wasi Ahmad
DDG [12] has published a paper explaining the design of

123
Int. j. inf. tecnol.

VoLTE, multiple voice transmission scenarios on LTE is therefore implemented in VoLTE by using eUTRAN and
network, the structure of the VOTE call and the challenges EPC, while voice resides in IMS.
associated with VoLTE, and has analyzed the efficiency of
VoLTE in different radio environments and voice quality 3.1 Mean opinion score (MOS)
in terms of MOS.Ayman Elnashar et al. [13]. The study
examined the use of VoLTE identification system to catch Mean opinion rating (MOS) is a metric that measures an
voice quality on VoLTE. intervention or system’s overall quality of performance of
Mohamed EL Wakiel and others [14] have measured knowledge or telecommunications technology. In all the
QoS KPIs by evaluating the trace of the actual VoLTE specific ‘‘values in the preset scale,’’ it is the numerical
calls, which they have registered, and verified output by measure that an object offers its opinion of a process
comparting them to the simulated values. quality’s performance [14], but they can also be algorith-
Guan-Hua Tuy et al. [15] analyzed VoLTE with respect mically calculated. Such scores are usually obtained in a
to cost-setting and mission complexity. VoLTE requires qualitative quality evaluation.
higher priority systems in mobile networks to ensure reli- MOS is a common measure used, but not limited to
ability. M also given the different conditions for the those methodologies, for video, audio and audio visual
delivery of voice call and retention services over LTE- quality assessment as shown in Table 1. ITU-T defined
based networks and a possible guidance for mobile oper- several ways in which the Score was taken from audiovi-
ators to maintain calling over LTE. M performed a test of sual, conversational, audio, speech and or video quality
practical VoLTE and IMS over LTE mobile reference testing, as defined in Recommendation P.800.1.
networks. Tabany et al. [16] The QoS output was analyzed
using OPNET and checked. The findings of the test were in 3.2 Mathematical rating scale
accordance with VoLTE ITU-R and 3GPP specifications.
The MOS is defined as a single rational number, typical of
1–5, where 1 is of the lowest quality and 5 of the highest
3 Voice quality perceived. Depending on the rating scale used for
traffic transmission methods in LTE networks the underlying test, other MOS ranges are also possible.
Quite common use is the Total Class Rating Scale indi-
The two different voice communication methods in this cated in Table 2 which maps rating from bad to outstand-
case are Simultaneous Voice or LTE, SVLTE and Circuit ing from 1 to 5 as per se.
switched back (CS FB) [17], enabling a mobile to con- ITU-T recommendations (such as P.800 or P.910) con-
currently utilize voice and data networks, but not to use tain other standardized quality scales.
them. When the whole networks are LTE, but they have no
facilities for making voice calls on the LTE, the voice calls PN
n¼1 Rn
are done by traditional networks such as CDMA/UMTS MOS ¼ ð1Þ
(2G/3 G). SVLTE utilizes two separate Radios to link N
Legacy CDMA to switched circuit services such as voice For instance, a constant scale between 1 and 100 could be
call and LTE network for improved packet switched server used. The scale of the test is dependent on the objective of
(PS) quality in the event of LTE issues with its legacy the trial. In some contexts, when obtained with different
CDMA network. [18] The CSFB is another potential LTE scales, there are no statistically significative differences
speech communication midway fix. In this situation, only between ratings of the same stimuli [21].
the existing networks are liable for a voice call made on the
LTE network. The device ‘‘goes back’’ to the 3 G or 2 G
network once LTE device is used for voice call or text Table 1 Satisfaction level of MOS score for voice call
message. Despite this, UE activates again on the LTE MOS Score Quality Level
network after the call is safe. [19, 20] LTE’s network and
VoLTE’s implementation was managed by LTE’s voice 4.3–5 Excellent Very satisfied
calls. Instead, if there is not LTE coverage, the existing 4–4.3 Good Well Satisfied
networks handle the calls. Single Radio Voice Call Con- 3.6–4 Regular Many satisfied users
tinuity (SRVCC), for voice traffic communication, is used 3.1–3.6 Poor Many dissatisfied users
for areas of non-LTE coverage. 2.6–3.1 Insufficient Almost all dissatisfied users
IMS puts together speech functions such as encryption, Less than 2.6 Bad Not recommended
database activation, call management, filtering, interoper-
ability with PSTN, billing etc. The communication network

123
 Int. j. inf. tecnol.

Table 2 Mathematical ratings

Rating Label
between bad and excellent
5 Excellent
4 Good
3 Fair
2 Poor
1 Bad

4 Call procedure in VoLTE

Two separately-adopted 3GPP specifications are based on

the VOLTE Architecture. In this sequence, IMS is the
primary VoIP operator for the LTE network because it
Fig. 1 LTE network architecture with (out) VoLTE [1]
knows and recognizes that specific network arrangements
are necessary to promote voice communication. IMS was
than data services, includes both VoLTE and spoken sig-
originally included in UMTS 3GPP 5 and LTE as an
nals. [15] The VoLTE service applies to two subsystems
excellent voice transmission service in shared data packets,
within LTE networks. The first is the cornerstone of the
synthesized with IMS and LTE. A network ISP with a
IMS, which is designed to support IP and multimedia
specified QoS is a carrier which provides greater spectral
products. The data gateway was built to provide VoLTE
efficiency, higher power, lower latencies, and required
consumers and traditional mobile users, for instance, with
service reliability (QoS). LTE guarantees this smooth IP
real-time multimedia traffic (voice). The VoLTE client
connectivity. The network frequently configures and
provides the features of the call control session between the
launches networks, as per request, in addition to the stan-
network, the broadcast portal and the 4G gateway. The
dard carrier that was approved for UE membership in the
second is the device for the distribution of current packets.
LTE network. Such modern companies are classified as
The 4 G portal is its main component. The main function is
independent carriers.
to provide the motive system with PS connectivity. This 4
The two key porters for voice traffic are SIP signs,
G gateway facilitates VoLTE by sending packets to power
which are popular for network transactions for the UE
and information planes between the network and the IMS
server or VoLTE in a VoLTE call.
center. The 4G gateway also facilitates control functions
The QoS Class Identifier (QCI) is allocated to each
like IP address distribution, packet filters, and network
client by the LTE network. A guaranteed or non-guaran-
quality and load assistance. The functional blocks for UE
teed bit-rate is considered by source type for each QCI. In
call control in case of ITRI VoLTE is shown in Fig. 2.
tandem with the radio tools and packet information
streams, QCI monitor the bearer from the UE to the PDN.
In the 3GPP standard TS 23.203 standard QCIs are
approved.
In the VoLTE system, the IMS framework sends guid-
ance to the LTE network using the Sip and the QCI sets
voice delegated cellular links to acceptable QoS facilities.
At the conclusion of the call, the IMS advises the LTE
network to minimize the actual Voice area. Of different
traffic forms, the QCI sets the appropriate latency rate.
The two communication sessions of each VoLTE call
are, according to Fig. 1, held on the data plane and on the
control plane. The control aircraft session is the exchanging
of call signals via the famous SIP. The data plane session
handles transmission of voice packets via the RTP, which
is defined at the inspection aircraft session on demand. To
maintain a similar call quality to standard CS calls, LTE
provides many product forms (for example, the fixed bit
rate and numerous priorities). The LTE data stream, which
offers standard data commodities but has higher emphasis Fig. 2 Functional for UE call control ITRI VoLTE [22]

123
Int. j. inf. tecnol.

The codec specified for VoLTE is an adaptive 3GPP Table 3 Video quality KPIs for
Video Observed frame rate
(AMR) codec. The use of AMR codec for VoLTE provides packetisation mode SNU (0)
formats (average)
benefits of interoperability of heritage networks.
QCIF 14.07
QVGA 14.01
VGA 13.78
5 Image and audio key performance indicators
(KPIs)

Image reliability has been tested in FU and SNU file for-

mats such as QCIF, QVGA, and VGA display. ‘‘KPI’’ Table 4 Video quality KPIs for
Video Observed frame rate
includes a description of the video capture(s) performance packetisation mode FU (1)
formats (average)
data. The below is the way to define KPIs. Frame rate: This
refers to the median number of framerates recorded in a QCIF 14
video session in all sections. Max Section- Segment: The QVGA 14.19
total framerate quality in a section in each session. Min VGA 14.09
section: The total framerate quality found in a line in a
session from every device. Frame rate Deviation Standard:
shows the framerates distribution across different seg-
ments. Normal Relative Deviation: this is the frame max- in Table 4 is the lowest measured frame rate of QVGA
imum relation to the frame rate observed. screen format. The frame rate is the maximum observed.
Time freeze: When two following frames are identical, it KPIs are assessed for QCIF, QVGA, and VGA display
is called frozen. Frozen one. The Time Frozen ratio rep- formats and for FU and SNU packet modes. Table 5
resents the time for which freeze happens during the ses- assesses and displays the KPIs.
sion. Percentage of time affected: a broken picture is an The session settings (average score) are calculated from
object that cannot be marked with any anomalies or cir- the measurements taken, which are the arithmetical mean
cular markings. The interpretation is demonstrated by the of the session values for all segments in a session. The
amount of time deficits in a session for which the period is weighted session rating is 4.97, excellently with a frame
affected. Session score: Based on the frame rate observed, rate of 15fps observed, as shown in Table 5.
frozen time percentage and time deterioration percentage.
The total session score is the numerical sum for all sections 6.2 Video quality investigation
of one session. Audio performance evaluation is similar to
MOS. Audio quality assessment. You can score every The Relative Signal Received Power (RSRP) picture
session at 5, 5 are nice, 4 are great, and 3 are average. quality control reaches -85 dB and the Noise and Interfere
KPIs are the middle view score for audio call which is Ratio (SINR) approaches 16. The object call of Codec
used to assess audio quality, as shown in Table 1 H.264 was tested, with a frame rate of 15 min in the SNU
unintentionally. and FU range of different video styles. For all test cases,
Table 6 shows the frame rate and session scoring.
Video Codec H.264 is scored 4.3 in the QCIF and SNU
6 Results and discussions features. Video codec H.264 in QVGA and FU formats
hold the highest 4.45 score with a 14.19 frame rate. Also in
KPIs for different display types and packeting modes the H.264 Video Codec, the best score is 14.09 in VGA-
determine the quality of video calls in VoLTE. and FU mode in session. The H.264 video codec was
evaluated in default Radio setters (RSRPs bigger than
6.1 KPIs for Video quality with distinct video - 110 dB and smaller than - 85 dB) for QCIF, QVGA,
formats VGA, SNU and FU system designs.
The graphical analysis is focused on graphs of averages
The KPIs for the SNU (0) mode were statistical tests of for all combinations. Image (Figs. 3 and 4) describe the FU
video quality. As shown in Table 3, the observed frame and SNU function for QCIF video rating distribution.
rate (average). For QCIF screen format, maximum is the YouTube score scale of 17 million ratings from 4.1 to 4.2,
measured frame rate (percentage). as shown in the Chart. In both QCIF and FU kit formats.
The video quality KPIs for the FU (1) mode are being The QCIF monitor size and SNU packeting mode shown in
quantitatively analysed. The median frame rate, as shown the statistic is 25 million quality at image score level

123
 Int. j. inf. tecnol.

Table 5 Video KPIs for different video formats

Key performance indicators Display Format QCIF QVGA VGA Expected Video
Quality
Pocketing station mode FU-A SNU FU-A SNU FU-A SNU

Observed frame rate (average) 14.00 14.07 14.19 14.01 14.09 13.78 15.00
Max segment 15.82 17.18 17.19 16.47 16.28 16.15 15.01
Frame rate standard deviation 1.52 1.79 2.04 1.70 1.86 1.78 0.01
Relative standard deviation 0.11 0.13 0.14 0.12 0.13 0.13 0.00
Target frame rate for score 15.00 15.00 15.00 15.00 15.00 15.00 15.00
Average session score (video quality) 4.24 4.30 4.45 4.41 4.46 4.43 4.97
Maximum session score (video quality) 4.89 4.90 4.89 4.89 4.83 4.88 5.00
Minimum session score (video quality) 3.51 3.44 3.82 3.76 3.76 3.64 4.90
Score standard deviation 0.40 0.44 0.36 0.36 0.38 0.37 0.03
Relative standard deviation 0.10 0.10 0.08 0.08 0.09 0.08 0.01

Table 6 Video quality test case

Sr. No. Test Case Packetization Mode Observed Frame Rate Session Score
results
1. H.264 SNU 14.07 4.3
QCIF
Frame rate set to 15
2. H.264 FU 14 4.22
QCIF
Frame rate set to 15
3. H.264 SNU 14.01 4.41
QVGA
Frame rate set to 15
4. H.264 FU 14.19 4.45
QVGA
Frame rate set to 15
5. H.264 SNU 13.78 4.43
VGA
Frame rate set to 15
6. H.264 FU 14.09 4.46

4.5–4.6. For codec H images. 264, QCIF’s perception of 4.7–4.8 array of video scores indicates 29% of the display
SNU mode offers better quality for video score than FU layout and the configuration of a VGA kit, shown in the
packets. picture. 8. With Video Codec H.264, the VGA output in
The image (Figs. 5 and 6) show FU and SNU packet the SNU packet format provides much more quality than
mode image output screen QVGA. The video ranking the video rating in FU packet mode (Fig. 8).
range 4.7–4.8 indicates 25 percent ratings in the monitor The SNU and FU packaging modes are available for the
type and the FU packeting feature. 4. As shown in the H.264 video codec and the range of pixels 4.7–4.8. As
QVGA screen style and SNU mode, 17 percent of video shown in Table 6, test results are confirmed. With a cal-
rating ranges 4.6–4.7 and 3.9–4.0. The QVGA packet culated frame score of 14.09, the session-level is correct at
layout for Video Codec H.264 is higher than the video 4.46. In the video-codec format H.264 QVGA or FU, video
rating in SNU style of 5. is available throughout 4.7–4.8. As shown in Table 6, the
The transmission of object levels for VGA with FU and test results were checked. At 14.19 the frequency is suffi-
SNU indicates categories 7 and 8, respectively. VGA ciently at 4.45.
monitor and FU pocketing mode shown in the figure are in The video codec H.264 is delivered in high quality 4.5
the picture range 4.7–4.8 for 25% values in Fig. 7. The to 4.6 output in QCIF and SNU pocketing modes. As

123
Int. j. inf. tecnol.

Fig. 3 Display format: QCIF,

packetization mode: FU

Fig. 4 Display format: QCIF,

packetization mode: SNU

shown in Table 5, the outcomes of these test cases are 6.3 Analysis of voice quality
confirmed. At the frame rate determined by 14.07, the
session score is correctly sustained until 4.3. The perfor- The HD apparatus was used to compute MOS for a range
mance of Video call is compared with 2G/3G, CSFB and of applications and provides an objective MOS that uni-
ViLTE as shown in Table 7. The session score quality in directional characterizes audio quality [23]. Get the
this case is best. The quality of ViLTE service in multi- microphone’s audio input into the system and the audio
mode and multiband interworking scenarios is also stated output from the headphone jack for contrast and determine
for reference in Table 8. MOS on the basis of different measures, including unidi-
The video codec H.264 is predicted to have 15 FPS rectional lag and missed images.
frame rate as per IR.94. This model defaults to FPS 15 Test calls are made in conjunction with each RSRP rate
QCIF. During the study, there was a significant video for every request, and the MOS is measured every 10 s for
failure for 30 FPS. Consequently, for the test cycle, the a period of 5 min. Table 1 shows the level of satisfaction
frame rate was kept at 15 FPS. The QCIF-15 Fps specifies with the MOS call ranking.
in the horizontal and vertical range of 176 9 144 pixels. For System under Test (DUT), the MOS software uplink
Such details help to improve video quality for device and downlinking is taken. The MOS information are shown
providers on the IMS stack. as the MOS distribution is organized every time. Table 9

123
 Int. j. inf. tecnol.

Fig. 5 Display format: QVGA,

packetization mode: FU

Fig. 6 DISPLAY format:

QVGA, packetization
Mode:SNU

summarizes the conditions of several testing environments Figure 10. MOS-Distribution displays uplink, MOS-
that measure the quality of the VoLTE services using the Range 3.0–3.1 it is 10%, 2.9–3.0 it is 20% and 2.7–2.8
POLQA [24, 25] under different signal strengths in terms MOS-Range is 40%.The MOS scope is weak uplink. This
of reference signal received power (RSRP). also indicates that the 90.00% of DUT A MOS values are
Case 1: The MOS downlink and the DUT A uplinks are below 3.0. All distributions show that the DUT A system is
displayed in Table 10. under the poor MOS range.
The image representation is shown in Figs. 9 and 10. In Table 11, Downlink and Uplink Case 2: MOS for
MOS-Downlink distribution 11%, 56% and 22%, indicates DUT B is shown.
the distribution of MOS in levels 3.1–3.2, 3.2–3.3 and The image representation is shown in Figs. 11 and 12.
3.3–3.4. This MOS scope indicates that consumers do meet Downlink MOS distribution is shown at MOS sizes
the call quality. The value of 77.78% of DUT A is above 3.3–3.4, 3.2–3.3 and 3.1–3.2 with values 20%, 60%, and
quality of MOS 3.2 10%, respectively. The MOS range suggests the customer

123
Int. j. inf. tecnol.

Fig. 7 DISPLAY format:

VGA, packetization mode:FU

Fig. 8 DISPLAY format:

VGA, packetization mode:SNU

Table 7 Performance of Video call in 2G/3G, CSFB, ViLTE [22] Table 8 Quality of ViLTE service in multimode and multiband
interworking scenarios [22]
2G/3G CSFB ViLTE
Experiment UE Configuration Bandwidth
Data/voice CS/PS CS/PS PS/PS
Call set up time 5s 6 to 8 s 2s FTP scenario 1 Band 3 FDD and Band 38 TDD CA 35 MHz
Voice quality AMR-NB AMR-NB HD AMR-WB FTP scenario 2 Band 41 TDD with 3CC CA 60 MHz
Video quality 64 CSD 64 CSD SVGA ViLTE call Band 28 FDD 20 MHz
MOS 3.6 3.6 4.08 Band 41 TDD

CS circuit-switched; PS packet-switched; AMR-NB adaptive multi-

rate narrowband; HD high definition; AMR-WB adaptive multi-rate
wideband; CSD circuit switched data; SVGA super video graphics
array

123
 Int. j. inf. tecnol.

Table 9 Test comparison in VOLTE [23]

Jia et al. [1] Villluz et al. [24] Nguyen et al. ITRI tests
[25]

Type of UEs Commercial devices N/A Self-deployed Both commercial devices and self-deployed
UE UE
Core network N/A N/A Lab simulation Nokia, Ericsson
RSRP [- 95 dBm, [- 75 dBm, N/A [- 95 dBm, - 110 dBm]
- 130 dBm] - 120 dBm]
Audio codec N/A N/A AMR WB AMR NB and AMR WB
Algorithm POLQA POLQA WB- E model POLQA
Reported MOS value [1.5, 4.2] [1.19,4.16] [3.47,4.47] [2.2,4.2]
range

Table 10 Downlink and uplink

Parameter DL UL
MOS for DUT A
Average 3.26 2.79
SD 0.09 0.16
Maximum score 3.37 3.02
Count 9 10
% MOS greater than or equal to 3.2 77.78% 0.00%
% MOS less than 3.0 0.00% 90.00%
% MOS less than 2.0 0.00% 0.00%
Scoring algorithm POLQA POLQA

Fig. 9 Downlink MOS

distribution for DUTA

123
Int. j. inf. tecnol.

Fig. 10 Uplink MOS

distribution for DUTA

Table 11 Downlink and

Parameter DL UL
Uplink MOS for DUT B
Average 3.23 2.82
SD 0.10 0.17
Maximum score 3.36 2.99
Count 10 11
% MOS greater than or equal to 3.2 80.00% 0.00%
% MOS less than 3.0 0.00% 100.00%
% MOS less than 2.0 0.00% 0.00%
Scoring algorithm POLQA POLQA

is not happy about the voice call quality. The value of DUT 7 Conclusion
B is 80% above the quality of MOS 3.2.
Scales 2.9–3.0, 2.8–2.9 and 2.7–2.8 and uplink MOS Summarized the some points: When call parties are the
variance from 36 to 9% are shown in the values distribu- UE’s of various brands, the ARM-NB Mode 7’s voice
tions. This is a MOS uplink. It also indicates that 100% of quality is better than the UE’s of the same brands. ARM-
MOS values for a given DUT B are below 3.0. The DUT B WB Mode 8’s output is influenced by UEs from various
in the low MOS range on a standard system is shown in all brands and thus call calculations by paring UEs from dif-
distributions. ferent brands are proposed. The VoLTE call quality is
The results are compared with performance indicators assessed using POLQA-NB and POLQA-WB to supply
for various scenarios as shown in Table 12. [22]. more genes.
With the IMS system, the VoLTE application imple-
mented voice in LTE network. In the current scenario,
IMS-based voice is widely considered the better approach.

123
 Int. j. inf. tecnol.

Fig. 11 Downlink MOS

distribution for DUTB

Fig. 12 Uplink MOS

distribution for DUTB

Image KPIs refer to the capture capture perfor- KPIs for video calls. The calculation analysis concludes
mance(s) for the various screen and packeting modes for with verification of the video codec option in VoLTE
the study of video quality in VOLTE. Graphical analysis of request for screen formats and packeting modes.
the video scoring distributions validate the calculated QoS

123
Int. j. inf. tecnol.

Table 12 Performance indicators of various scenarios [23]

ARM Scenario Call parties POLQA score

ARM -NB mode 7 Scenario 1 Two UE A (Vendor A) Range [3.1, 3.9]; average is 3.54
Scenario 2 Two UE B (Vendor B) Range [3.0, 3.9]; average is 3.48
Scenario 3 Two ITRI VoLTE Tests Range [2.6, 3.9]; average is 3.42
ARM -NB mode 7 Scenario 4 UE A and UE B UE A Range [3, 3.9]; average is 3.57
UE B Range [3.3, 3.9]; average is 3.59
Scenario 5 UE B and ITRI UE UE B Range [3.1, 4.2]; average is 3.88
ITRI UE Range [1.9, 3]; Average is 2.34
ARM-WB mode 8 Scenario 6 Two UE A Range [3.1, 4.0]; average is 3.64
Scenario 7 Two UE B Range [3.0, 4.0]; average is 3.58
Scenario 8 UE A and UE B UE A Range [2.2, 3.1]; average is 2.79
UE B Range [3.0, 4.0]; average is 3.76

The Video Codec H.264 is a VGA with the SNU and FU 7. ITU-T (2014) Perceptual objective listening quality assessment;
option in the 4.7–4.8 video range. With a picture rate of ITU-T Recommendation, p 863
8. Sung Y-C et al (2012) NCTU-VT: a Freeware for Wireless VoIP
14.09 the H.264 video codec is held in a 4.7–4.8 score with Performance Measurement. Wirel Commun Mob Comput
a QVGA or FU packaging format. The photographic score 12(4):318–324
of 4.46 is correctly measured. The Rating is accurately held 9. GPP (2017) Technical Specification Group Services and System
at 4.45 with frame rate observed 14.19. The findings of a Aspects; ANSI-C code for the Adaptive Multi Rate (AMR)
speech codec; Technical Specification 3GPP TS 26.073 version
video test case were checked for video codec quality. 14.0.0 (2017-03)
The quantitative MOS scale referred for assessing voice 10. ITU-T (2015) Wideband E-model; ITU-T Recommendation
quality shall be conducted for voice call analyses. MOS G.107.1
DUT was shown by the Downlink/Uplink MOS graphical 11. GPP (2017) Evolved Universal Terrestrial Radio Access (E-
UTRA); Physical layer; Measurements, Technical Specification
projection. The DUT case study categorized it as weak 3GPP TS 36.214 version 14.1.0 (2017-01)
MOS phones potentially. 12. Wasi Ahmad DDG (2016) Study Paper on Voice over LTE: New
Voice Dynamics, 2016-2017
13. Elnashar A, El-Saidny MA, Yehia M (2018) Performance Eval-
uation of VoLTE Based on Field Measurement Data
14. Wakiel MEL, Badawy HEL, Elhennawy H (2016) Design of
References quality of service parameters for voice over long term evolution
‘‘LTE’’ network. Int J Sci Basic Appl Res (IJSBAR), ISSN
1. Jia YJ et al (2015) Performance characterization and call relia- 2307-4531
bility diagnosis support for voice over LTE. ACM. ISBN 978-1- 15. Tuy G-H, Liy C-Y, Pengz C, Yuany Z, Liy Y, Zhaoz X, Luy S
4503-3619-2/15/09 (2016) VoLTE*: A lightweight voice solution to 4G LTE net-
2. GPP (2015) Technical Specification Group Services and System works. February 26–27, 2016, St. Augustine, FL, USA, ISBN:
Aspects; IP Multimedia Subsystem (IMS); Stage 2; Technical 978-1-4503-4145-5. https://doi.org/10.1145/2873587.2873604
Specification 3GPP TS 23.228 version 13.3.0 (2015-06) 16. Tabany MR, Guy CG (2014) An end-to-end QoS performance
3. GPP on track to 5G. http://www.3gpp.org/newsevents/3gpp- evaluation of VoLTE in 4G E-UTRAN-based wireless networks.
news/1787-ontrack_5g ICWMC 2014: 10th International conference on wireless and
4. GPP (2016) Technical Specification Group Services and System mobile communications. ISBN: 978-1-61208-347-6
Aspects; Circuit Switched (CS) fallback in Evolved Packet Sys- 17. GPP TS 22.228 V12.9.0, Service requirements for the Internet
tem (EPS); Stage 2; Technical Specification 3GPP TS 23.272 Protocol (IP) Multimedia core network Subsystem (IMS)
version 13.4.0 (2016-06), 2016 18. GPP TS 22.173 V9.5.0 (2010-03) IP multimedia core network
5. GPP (2015) Evolved Universal Terrestrial Radio Access (E- subsystem (IMS) multimedia telephony service and supplemen-
UTRA); User Equipment (UE) radio transmission and reception, tary services; Stage 1 (Release 9)
Technical Specification 3GPP TS 36.101 version 13.0.0 (2015- 19. Elnashar A, El-Saidny MA (2013) Looking at LTE in practice: a
07) performance analysis of the LTE system based on field test
6. GPP (2017) Technical Specification Group Services and System results. IEEE Veh Technol Mag 8(3):81–92
Aspects; Internet Protocol (IP) multimedia call control protocol 20. Elnashar A, Al-Saidny M, Sherif M (2014) Design, deployment,
based on Session Initiation Protocol (SIP) and Session Descrip- and performance of 4G-LTE networks: a practical approach.
tion Protocol (SDP); User Equipment (UE) conformance speci- Wiley, Hoboken
fication; Part 3: Abstract test suite (ATS); Technical Specification 21. Tabany MR, Guy CG (2013) Performance analysis and deploy-
3GPP TS 34.229-3 version 13.1.0 (2017-01) ment of VoLTE mechanisms over 3GPP LTE-based networks. Int
J Comput Sci Telecommun 4(10):1

123
 Int. j. inf. tecnol.

22. Sung Y-C et al (2018) Voice/video quality measurement for LTE Proceedings of the International Conference Information and
services. IEEE Wirel Commun. https://doi.org/10.1109/mwc. Communication Technology Convergence (ICTC)
2018.1700109 25. Nguyen D-H, Nguyen H (2016) A dynamic rate adaptation
23. Li L et al (2017) A measurement study on skype voice and video algorithm using WB E-model for voice traffic over LTE network.
calls in LTE networks on high speed rails. 978-1-5386-2704-4/17 2016 Wireless Days (WD)
24. Villaluz MZ et al (2015) VoLTE SRVCC Optimization as Interim
Solution for LTE Networks with Coverage Discontinuity. In:

123