IJRET: International Journal of Research in Engineering and Technology

eISSN: 2319-1163 | pISSN: 2321-7308

PERFORMANCE ANALYSIS OF VoIP OVER WIRED AND WIRELESS

NETWORKS : NETWORK IMPLEMENTATION IN ADEN
UNIVERSITY

Mohsen Hussein Mohammed1, Wafaa Nasser Abdullah2

Associate Professor, Department of Information Technology, Engineering College, Aden, Yemen

Research Scholar, Department of Information Technology, Engineering College, Aden University, Aden, Yemen

Abstract

In this Paper, the objective of simulation models is presented to investigate the performance of VoIP codecs over WiMAX and
FDDI networks that specially design for Aden University. To assure if the University IP network is prepared and adequate for this
new type of traffic before adding any new components, Aden University IP network will be simulated by using OPNET simulation
software then the new VoIP service will be added to the University networks. Different parameters that represent the QoS like end
to end delay, jitter, traffic sends and traffic received, MOS are calculated and analyzed in both network scenarios.

Keywords: VoIP, Codecs, QoS, WiMAX, FDDI.

--------------------------------------------------------------------***--------------------------------------------------------------------1. INTRODUCTION

Through the last few years, there was a progression trend

into real-time voice communication using Internet protocol
(IP).This progress caused by the merger of voice and data
traffic (telecommunication convergence) over the existing
network infrastructure, improved network management and
lower cost offered by technology. VoIP (Voice over Internet
Protocol) is a communication technology that supports
delivery of voice over the Internet or other packet switched
networks rather than the Traditional Public Switched
Telephone Network (PSTN) [1]. The wired Internet Protocol
(IP) networks given higher VoIP services as compared with
the wireless IP network as wireless networks and they have
their own characteristics and weakness [2]. The unsolved
problems triggered via the wireless network in this place
still need a few dedicated work spotlighting VoIP calls [3].
In this project, the OPNET simulator is used to implement
the proposed VoIP network. We study all the potential
parameters that can decline the quality of VoIP over two
types of networks as FDDI and Wireless MAN (WiMAX)
designed for Aden University. The study is depended on the
choice of an appropriate voice compression and
decompression (CODEC) schemes depending on the Quality
of Service (QoS) of VoIP in variant networks. Different
parameters that indicate the QoS like end-to-end delay, jitter
throughput, and MOS, are calculated and analyzed. The
comparisons were carried out between different codecs
(G.711, G.729A, and G.723.1) which are the most
appropriate to improve QoS for VoIP.
The whole parts of this paper are organized as follows.
Some related works are mentioned in Part 2. Part 3 concerns
with VoIP codec schemes. The model design and
configuration networks are described in Part 4. Part 5

presents simulation results and analysis. Finally, Part6

concludes the paper.

2. LITERATURE REVIEW

Many authors have worked on diverse Quality of Service

(QoS) parameters using unique service classes in different
network types. A study was performed on various quality
parameters influencing on the VOIP service performance.
The study shows that these parameters of QoS are required
to increase VoIP performance. Gihad Yousif Gafar [4],
indicates that the simulation results have a significant effect
on the performance of VoIP in WiMAX networks ,which
refer that better choice of voice codec's and statistical
distribution and performance of selected parameters will be
done using the network simulator OPNET Modeler. In
Takahashi et al [5], the authors attempt to identify some of
the network performance parameters that Service Providers
will consciousness directly to develop a VoIP over WiMAX
communication tool. That serves as a voice communication
broadband alternative technology to old circuit-switched
voice communication. Abdul-Bary Raouf Suleiman et al.
[6], present a simulation for SIP-Based based VoIP in Mosul
college community. Voice telephony is taken into
consideration as a real-time service carrier and time
sensitive service. The performance of network and voice
quality examine after including the VoIP for distinct codec
types. The results are presented that Mosul University IP
network is successful to enforce the VoIP. Hussein et al, [7],
linked a comparison between different queue algorithms.
The authors found that Priority queue and Weight Fair
Queuing algorithms are the most appropriate to improve
QoS for VoIP. R. Nandhini et al. [8], represent that the
objective of the broadband wireless technologies is to assure
the end-to-end Quality of Service for the classes of service.
The proposed algorithm describes in detail, and evaluates

_______________________________________________________________________________________
Volume: 05 Issue: 02 | Feb-2016, Available @ http://www.ijret.org

325

IJRET: International Journal of Research in Engineering and Technology

with real-time video traffic and one VOIP codec, by a series
of simulations involved in this interesting location of
studies. In [9] Ali M.Alsahlany, evaluated the VoIP
performance based wireless LAN/WAN network with
taking into account different voice encoding schemes.
Different parameters that refer the QoS like jitter, end-toend delay, traffic sends, traffic receives and MOS measured
and analyzed in Wireless LAN /WAN scenarios. However,
all the previous studies are designed in the literature based
on either wired or wireless network or both. Most of them
designed for LAN or WAN networks. Therefore, it is
essential to do a research on both wired and wireless MAN
networks from different aspects. This paper examines VoIP
performance of campus; MAN network, containing both
wired and wireless network components. The results of this
paper can help organizations make decisions for adopting
VoIP system and expansion plans for VoIP services. A
number of studies have published where the performance of
several codecs evaluate and compare under different
conditions such as the performance of codec on a Local
Area Network (LAN) and wide area network (WAN) with
different characteristics such as a number of users and
different networks.

3. VOIP CODECS

Voice Codecs are used on the client side to convert the

analog voice signal to digital signal and vice versa. There
are various codecs types depended on the selected data rate,
sampling rate, and implemented a compression algorithm
listed in Table.1. In order to locate the bandwidth
requirements of VoIP connections, we will begin by
presenting common VoIP Codecs and their associated
characteristics[10].
G.711 is currently used in a wide domain of applications. It
employs a logarithmic compression that compresses each
16-bit sample to 8-bits. As a result, it digitizes voice into 64
kbps, which consider the highest bit-rate among the codecs.

eISSN: 2319-1163 | pISSN: 2321-7308

It performs best in local networks where we hold lots of

available bandwidths[10],[11].
G.729 is an authorized codec designed to deliver good call
quality without exhaustion of high bandwidth. It builds
based, on the Conjugate-Structure Algebraic-Code-Excited
Linear Prediction (CS-ACELP) algorithm, with a bit rate of
8 kbps[11].
G.723 is also an authorized codec. It is designed for calls
over modem links with data rates of 28.8 and 33 kbps. It
operates at 6.3 and 5.3 Kbps. Although this standard
decreases bandwidth exhaustion, the voice is much poorer
than with G.729 and is not very common for VoIP [11].
Table -1: VoIP Codec Characteristics
Packet
Samplin
CODEC
Coding Algo
Per
g rate
Second
G.711
PCM
64 kbps
100
G.723.1
ACELP
5.3 kbps
33
G.729A
CS-ACELP
8 kbps
100

IP
Packet
Size
120
60
50

5. NETWORK MODEL DESCRIPTION

Our simulation approach uses OPNET simulator for network

modeling. OPNET simulator is an authoritative
communication system simulator developed by OPNET
technologies. Two simulation models were to determine the
best audio encoding schemes of utilizing VoIP over both
WiMAX and FDDI networks. Both networks follow a
different structure and the same topology. The basic
difference in all the scenarios was in the configuration of the
corresponding application and the profiles. Each scenario is
implemented with the codec G.711, G.723, and G.729.
Various comparisons are conducted to find the values of
various parameters. The simulation models for the networks
under study are illustrated in Figure 1 (a) and Figure 1 (b).

Fig -1(a): FDDI Network

_______________________________________________________________________________________
Volume: 05 Issue: 02 | Feb-2016, Available @ http://www.ijret.org

326

IJRET: International Journal of Research in Engineering and Technology

eISSN: 2319-1163 | pISSN: 2321-7308

Fig -1(b): WiMAX Network

The Application Config, Profile Config, Mobility Config
and QoS Attribute Config are included in the models. The
experimental network models consist of Wired and Wireless
network setup in the University campus. The wired model
consists of two main routers, which are connected to eight
different parts of campus by fiber optic links and each zone
has an equal number of VoIP users. On the other side,
Wireless network is similar to the wired network, where
LANs in all the eight zones are the same, the connection
between the center and each zone is replaced by wireless
routers on the other hand. In a wireless network, we
implemented SIP proxy server (session initiation protocol)
to work together with FTP (file transfer protocol) and HTTP
(Hyper Transfer Text protocol) as VoIP server.

5.1 Basic Parameter

Wireless MAN environment uses 1Gbps Ethernet cable for

all LAN networks. The OPNET model name for servers is
'ethernet_server'. The wireless MAN environment contains 1
wireless Router and 2 fixed workstations. All these nodes
use VoIP services. There are a total of 16 workstations in
each Wireless MAN environment. The router Opnet model
names are 'WiMAX_ss_ethernet_slip_WLAN_router_adv'
as WiMAX subscriber station routers and 'WiMAX_bs_ethernet4_slip4_router' as WiMAX base station
routers. The fixed VoIP workstation model name is
'1000BaseX_LAN'. The workstations are generating traffic
across the LANs and the Wireless MAN environment to
simulate a campus office environment.
Different types of connection connect all University
buildings. With Wired network, fiber optic cables distribute
around the University and connecting all campus buildings
with the router Opnet model 'ethernet_fddi_gtwy_base'.
Opnet modeler 14.0 is used as simulation software for the
university campus network [12]. All parameters used in the
wired and wireless MAN environment show in Table 2.

Table -2: Parameters Used In Simulation

Parameter

Value

Workstation

1000BaseX_LAN

Switch

eth16_ethch16_fddi16_tr16_switch

WiMAX_bs
_router

wimax_bs_ethernet_slip_wlan_router_adv

Server

Router

WiMAX_ss
_router
Links

ethernet_server

ethernet_fddi_gtwy_base

wimax_ss_ethernet_slip_wlan_router_adv
- FDDI 100MB

-1000BaseX duplex link

5.2 Application, Profile and Configurations

Application Definitions are used to create traffic generation

on the network. There are four applications utilized in each
scenario. Those are VoIP, FTP, web browsing , and database
applications. The essential one is VoIP. Within the Voice
category, IP Telephony was selected for configuration VoIP.
Within the VoIP application, voice attributes were set-up as
codec schemes G.711 for scenario 1; G.723.1 for scenario 2;
and G.729A for scenario 3. G.729A runs at a bit rate of 8
Kbps, G.723.1 runs at a bit rate of 5.3/6.3 Kbps and G.711
runs at a bit rate of 64 Kbps. For the QoS scenarios, the
VoIP table type of service (ToS) is changed into six to offer
voice priority. Different parameters that have been set for
this venture were internet browsing and file transfer. For
each of these applications stated, there has been a heavy
usage parameter set.

_______________________________________________________________________________________
Volume: 05 Issue: 02 | Feb-2016, Available @ http://www.ijret.org

327

IJRET: International Journal of Research in Engineering and Technology

eISSN: 2319-1163 | pISSN: 2321-7308

Profile Configuration describes the activity applications

used by users through the duration. VoIP_Profile was only
used with VoIP applications that were the committed VoIP
phones. All of the Profiles configured to run simultaneously
to permit more than one application to be operated at the
same time.

6. SIMULATION RESULTS AND ANALYSIS

The following figures are obtained by accumulating

statistics by using the OPNET Modeler Simulation tool. All
figures show a comparative picture of the 3 scenarios. All of
the 3 scenarios in each network use a specific audio codec
scheme consisting of three codecs G.711, G.723 and G.729.
After running the simulation successfully, the final result
suggests the impact of various codecs on different QoS
parameters in a VoIP network. The following are the figures
shown below display different QoS parameters like packet
end to end delay, Voice jitter, Voice traffic sends
(packet/Sec) and Voice traffic receives (packet/Sec), and
MOS.

End to End Delay

The delay is the amount of time that takes a packet from one
end of the network to the other. Unlike many data
applications, voice can't afford high levels of latency or
delay. Therefore, for any network to be more efficient, it
must have less delay value. It is calculated in terms of sec or
min. The Following graphs of delay shown in figure 3 (a)
and 2 (b) are drawn between the simulation time and basic
time, which is termed as the delay of the network from both
graphs. G711 has less delay with compared to the others.
Therefore, we can say that these networks are more efficient
in this delay term in wired network. In figure 2(a) both
G.711 and G.29 have the same delay value and this value is
less than the other network, while a voice delay of the G.723
codec scheme becomes worse. In figure 2(b), G.711 has less
delay value than the other network scenarios. From both
graphs, G.711 is much better in WiMAX and G.711and
G.29 are much better in wired networks.

Jitter

Fig -2(b): WiMAX (Voice ETE Delay)

Jitter is the value, which represents the time taken by

packets to reach the destination. This variation is caused due
to route changes, timing drift or network congestion. It is
calculated in terms of Sec or minute. Perceived voice quality
is the best if the jitter is zero. These graphs are drawn
between the simulation time and basic time, which is termed
as jitter of the network. In figure 3(a) graph, the jitter of
G.711, G.723, and G.729 decrease and become very close to
zero. In figure 3(b) graph, G.711 has the best jitter value and
this value less than other network scenarios. Therefore,
these networks are better in this jitter term. From both
graphs, G.711 is much better.

Fig -3 (a): FDDI: Voice Jitter

Fig -2 (a): FDDI (Voice ETE Delay)

_______________________________________________________________________________________
Volume: 05 Issue: 02 | Feb-2016, Available @ http://www.ijret.org

328

IJRET: International Journal of Research in Engineering and Technology

eISSN: 2319-1163 | pISSN: 2321-7308

Fig. 3(b): WiMAX (Voice Jitter)

Throughput

Figure 4(a) and 4(b) are the graphs of traffic throughput

in both figures; two traffics must be the same for any
network to be more efficient. In Figure 4(a) network
scenarios indicate that the G.729 scenario is the best in
traffic send and receive in comparison with other scenarios.
Otherwise, Figure 4(b) scenarios, traffic received by the
network with G.711 is more deviated from the traffic sent,
whereas the G.729 network, the deviation is much less. This
analysis indicates that the noise added in the G.729 scenario
is less with compared to other scenarios. Therefore, we can
say that this network is more efficient.

Fig -4(b): WiMAX (Voice Throughput)

Mean Opinion Score (MOS)

MOS (Mean opinion score) In the VoIP, MOS is a number

between 1 and 5 used to describe the voice perception
quality in communication systems, especially digital
networks that transfer VoIP traffic. Therefore, Voice quality
measures by the Mean Opinion Score (MOS). Figures 5(a)
and 5(b) display the MOS factors of our networks for three
Codec types. We observe that the MOS factors for Codecs
G. 711 and G. 729 have acceptable MOS values 3.685 and
3.067, respectively. On the other hand, the MOS values for
G. 723 are 2.557 in FDDI network and 1 in WiMAX
network, which indicates that the quality of service is poor
and unacceptable if this codec is used.

Fig -4(a): FDDI (Voice Throughput)

Fig -5(a): FDDI (Voice MOS Value)

_______________________________________________________________________________________
Volume: 05 Issue: 02 | Feb-2016, Available @ http://www.ijret.org

329

IJRET: International Journal of Research in Engineering and Technology

Fig -5(b): WiMAX (Voice MOS Value)

7. CONCLUSIONS

In this implementation, we evaluated the performance of

VoIP codecs over both FDDI and WiMAX networks. Three
voice codecs G.711, G.723, and G.729 simulated in order to
determine the most appropriate voice codec for VoIP over
FDDI and WiMAX network. The VoIP performance
simulate via six simulation scenarios by using OPNET 14.0.
Different parameters such as end-to-end delay, jitter,
throughput, and MOS, are used to measure VoIP
performance. The implementation results of this paper show
that the VoIP services perform best under G.711 voice
encoder scheme over WiMAX network. Otherwise, in FDDI
network, VoIP services perform best under both G.711 and
G.729 voice encoders schema. In addition, in FDDI network
voice throughput is higher in codec G.729 as compared to
the other two referenced codecs.
The future research can be done in advance wireless
standard and equipment for WiMAX network by
implementing the network in the latest edition of OPNET
MODELLER 17. In addition, calculating the performance of
codec includes the combination between more than one
codec to achieve a VOIP application using different
networks. FDDI and WiMAX networks may also use other
VoIP codecs such as G. 722, G. 726, GSM, and G.728.

REFERENCES

[1]. S. Dhanalakshmi, and M. Sathiya, Investigating The

Performance Of VoIP Over Ethernet LAN In Campus
Network International Journal of Recent Scientific
Research,Vol. 6, Issue, 6, pp.4389-4394, June, 2015 .
[2]. A.Samukic,"Umts Universal Mobile Telecommunications System Development of Standards for the Third
Generation, 1998.
[3]. Malik Ahsan Ali, Imran Rashid, and Adnan Ahmed
Khan, Selec-ion of VoIP CODECs for Different Networks
based on QoS Analysis International Journal of Computer
Applications (0975 8887) Volume 84 No 5, December
2013.

eISSN: 2319-1163 | pISSN: 2321-7308

[4]. Gihad Yousif Gafar, Enhanced Quality of Service in

Worldwide Interoperability for Microwave Access Networks, IOSR Journal of Electronics and Communication.
Engineering (IOSR-JECE), Volume 9, Issue 5, Ver. V (Sep Oct. 2014), PP 72-78.
[5]. Elechi Onyekachi O.1, and Eze Elias C.1*,
Investigating the QoS of Voice over IP using WiMAX
Access Networks in a Campus Network, Computer,
Engineering and Intelligent Systems, ISSN 2222-1719
(Paper) ISSN 2222-2863 (Online), Vol.4, No.5, 2013.
[6]. Abdul-Bary Raouf Suleiman & Abdul-Hameed Hameed
Simulation of SIP-Based VoIP for Mosul University
Communication Network, International Journal of
Computing and Digital Systems, Int. J. Com. Dig. Sys.2,
No. 2, 89-94(2013).
[7]. Hussein, and et al, The Effects of Different Queuing
Algorithms within the Router on QoS VoIP application
Using OPNET,International Journal of Computer Networks
& Communications (IJCNC) Vol.5, No.1, Jan, pp. 117-124,
2013.
[8]. R.Nandhini, N. Devarajan, Channel quality based
cross-layer scheduling algorithm in Wimax networks
available online: http://www.lifesciencesite.com, Life
Science Journal 2013; 10(2).
[9]. Ali M. Alsahlany, Performance Analysis of VoIP
Traffic Over integrating Wireless LAN and WAN Using
Different Codecs, International Journal of Wireless &
Mobile Networks (IJWMN) Vol. 6, No. 3, June 2014.
[10]. Tarik Anouari and Abdelkrim Haqiq, Performance
Analysis of VoIP Traffic in WiMAX using various Service
Classes, International Journal of Computer Applications
(0975 8887) ,Volume 52 No.20, August 2012.
[11]. Gysberth Maurits Wattimena, Analysis Performance
VoIP Codecs over WiMAX Access Network ,International
Journal of Advanced Research in Computer Science and
Electronics Engineering Volume 1, Issue 7, September 2012
[12].OPNET,http.//www.telecomlab.oulu.fi/kurssit/521365
A_tietoliikennetekniikan_simuloInnit_ja_tyokalut/Opnet_es
ittely_07.pdf, journal, Retrieved. 2013-12-05.

BIOGRAPHIES

Mohsen Hussein Mohammed associate professor, vice

Dean of students Affairs in Faculty of Computer and
Information Technology Aden University- Yemen. He has
received the PhD degree in Computer Systems, Complexes
and Networks from the Faculty of Engineering Blagoevgrad
University-Bulgaria by 2006. He has published a number of
researches in various journals and international conferences.
His current research interests are in parallel computer
architecture, Multicomputer platform, System area
Networks. E-mail: [email protected]

Wafaa Nasser Abdullah research scholar, she has

worked as a lecturer in the Computer Department at
Zanzibar Education College Aden University since 2010
till now. She received the B.S of Engineering degree from
Department of Information Technology - Faculty of
Engineering- Aden University, Yemen, in 2009, Her current
research interests are in VOIP over WiMAX network.
E-mail: [email protected]

_______________________________________________________________________________________
Volume: 05 Issue: 02 | Feb-2016, Available @ http://www.ijret.org

330