See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/324144856

Performance Evaluation of Wireless 802.11n Using Level 2 and Level 3 Mobility

Article in Indian Journal of Science and Technology · April 2018

DOI: 10.17485/ijst/2018/v11i14/120616

CITATIONS READS
6 621

3 authors:

Hind Sounni Najib EL KAMOUN

Université Chouaib Doukkali Université Chouaib Doukkali
9 PUBLICATIONS 15 CITATIONS 189 PUBLICATIONS 548 CITATIONS

SEE PROFILE SEE PROFILE

Fatima Lakrami
Université Chouaib Doukkali
30 PUBLICATIONS 64 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

MIGRID Project View project

a study of SD-WAN vs Traditional WAN Architecture, Comparison and performance analytics View project

All content following this page was uploaded by Hind Sounni on 19 April 2018.

The user has requested enhancement of the downloaded file.

 ISSN (Print) : 0974-6846
ISSN (Online) : 0974-5645
Indian Journal of Science and Technology, Vol 11(14), DOI: 10.17485/ijst/2018/v11i14/120616, April 2018

Performance Evaluation of Wireless 802.11n Using

Level 2 and Level 3 Mobility
Hind Sounni*, Najib Elkamoun and Fatima Lakrami
Department of Physics, STIC Laboratory, Chouaib Doukkali University, El Jadida – 24000, Morocco;
[email protected], [email protected], [email protected]

Abstract
Objectives: To present a performance evaluation of a WI-FI network under the impact of mobility. It alsomakes
the study and the evaluation cover for level 2 and level 3 mobility. Methods/Statistical Analysis: The OPNET
simulator was used for simulating the scenarios of proposed architectures. For performance evaluations, the dif-
ferent parameter has considered, such as mobility speed, number of connected users and channel access methods.
Findings: The main interest of this paper is to reveal the mobility schema that ensures a better continuity of provid-
ing services to mobile users moving between different Wi-Fi access points while considering a number of constraints
related to the quality of service. Application/Improvements: Deploying level 2 roaming gives better results com-
pared with level three roaming using MIPv4, while increasing the speed of mobility and the number of clients.

Keywords: Level 2 Mobility, Level 3 Mobility, Performance Evaluation, MIPv4, 802.11 n

1. Introduction ity using Mobile IPv4 technology by bringing different

application schemes that can be used in a network. The
The number of wireless devices for voice or data exceeds rest of the paper has organized as follow: section 2 gives
the number of fixed devices. Mobile data communication a description of wireless network parameters. Section 3
will likely emerge as the technology supporting most com- describes the mobility in a wireless network. The per-
munication types including voice and video. Mobile data formance evaluation is presented by section 4. Section 5
communication is present in cellular systems such as 3G resumes simulation results, while section 5 concludes the
and in wireless LAN such as 802.11 and will extend into paper.
satellite communication. Wireless fidelity or Wi-Fi is one
of the most used types that makes connection costs for
large networks easier, cheaper, and allows wide coverage
2. Wireless Network Parameters
while assuring the QoS requested by users using multi- In wireless networks, several parameters are configurable
media applications. Though mobility may be enabling and have a huge impact on wireless network performance
link-layer technologies, data crossing networks or dif- (Wi-Fi), such us:
ferent link layers is still a problem. The solution to this 1. Access point properties: Antenna Gain,
problem is a standards-based protocol, Mobile IP. This Modulation, Wireless Card, Standard, and
paper provides an overview of level 2 and level 3 mobil- Frequency, Transmit Power and Coverage,

*Author for correspondence

 Performance Evaluation of Wireless 802.11n Using Level 2 and Level 3 Mobility

2. Number of users: Channel access competition, subnet1 (APs belonging to the same subnet) Figure 1. The
roaming includes a movement; it is quite conceivable to
3. Mobility of users: Specifies whether the mobile imagine during a movement to end up on another IP net-
node should switch to access points with higher work than the original one. If the terminal connects to
signal strength either by using the level 2 mobility, an AP in a different subnet, then network (L3) roaming
and level 3 mobility,
will happen. This level of mobility has known as Macro-
mobility Figure 1. Actually, Network roaming (L3) can
4. Load: Based on the number of packets in network,
occur only after a successful L2 roaming1,2.
and

3.1 Mobility
5. Type of traffic: Best effort or Multimedia.
Micro-mobility is the movement of a mobile node
This paper is particularly interested about studying between two attachment points on the same network 2.
the impact of two factors: mobility in term of speedand The basic features of a mobility management are mainly
the number of users. the management of the location and management of
handovers. For that, some functions must necessarily
have a protocol authorizing the mobility management,
3. Mobility in Wireless Network
citing the transfer of the packets, the update of the ways or
There are several levels of mobility referring to the OSI the roads, the management of the handovers, the support
model. The lowest level of mobility is level 2, which cor- of the inactive mobiles, the management of the addresses,
responds to the link layer of the OSI model. This level the authentication and security mechanisms, etc. The cri-
of mobility has known as Micro-mobility. The talking is teria for selection of a mobility management protocol are
about data-link (L2) roaming when a terminal re-associ- mainly the total duration of the relay and packet loss rate
during the procedure. This called “Smooth Handover” to
ates with a new AP from the current AP within the same
designate a relay with a minimum of delay.

Figure 1. Layer 2 and layer 3 roaming.

2 Vol 11 (14) | April 2018 | www.indjst.org Indian Journal of Science and Technology
 Hind Sounni, Najib Elkamoun and Fatima Lakrami

3.2 Macro-Mobility • FA: (Foreign Agent) is a router on the FN, and

Macro-mobility is about moving between two different
• CN: (Correspondent Node) is a node that has
domains (networks). It can take place during an active
intended to communicate with a MN. It may be
session of a mobile user, or during a new session initiated
mobile or a stationary node5.
by the user in a visited network. Mobile IP has designed
to handle this type of mobility2.
Mobile IP uses the technique of re-routing to allow
3.3 Mobile IP Overview mobile nodes to roam around wireless LANs without dis-
rupting transport sessions6. By designating a Home Agent
Traditional IP technology cannot support mobility in IP
or HA on the mobile home network which must intercept
layer.When a mobile node moves and attaches itself to
the packets intended for the mobile and then encapsu-
another network; it needs to obtain a new IP addressthe
late them to be sent to the foreign agent FA located in
reason why IETF defines Mobile IP3. The fundamen-
the visited network that intercepts packets destined for
tal assumption behind the development of Mobile IP
MN within the FN, encapsulates them and finally delivers
was backward compatibility with the existing Internet
them to the MN7. When MN is on its home subnet, it uses
infrastructure based on TCP/IP protocol suite originally
normal IP protocol to communicate. While it enters the
developed for fixed networks4.
foreign subnet, it acquires an IP address, Called Care Of
Figure 2 describes the functioning of mobile IP net-
Address (COA). It then sends a registration message to
work. Different entities used in mobile IP networkare:
HA to inform HA of its current location, COA8. When a
correspondent CN decide to rejoin the mobile node, and
• MN: (Mobile Node) is a node that shows mobility
it sent a datagram, which reaches the home network via
without changing its IP i.e. it can change its point
standard IP routing. The datagram is intercepted by home
of attachment from one link to another but will be
agent HA and is tunneled to the care-of address, then
reachable through its home address,
delivered to the MN.
Each time the mobile node enters a new network, a
• HA: (Home Agent) is a router on the HN that pro-
handoff triggers. The mobile gets a new temporary COA
vides services to MN,

Figure 2. Mobile IP network.

Vol 11 (14) | April 2018 | www.indjst.org Indian Journal of Science and Technology 3
 Performance Evaluation of Wireless 802.11n Using Level 2 and Level 3 Mobility

address and registers it to HA again in order to be able OPNET 17.5 is used to evaluate the performance of the
to forward the datagram to it. When the MN leaves the proposed network, which makes it possible to build simu-
old foreign network the mobile node sends registration lations of network infrastructure. In this simulation, the
request message to it’s HA, indicating the change of its access points use 802.11n standard9 (Data Rate 54 Mbps).
position and inform the HA to stop sending datagrams to Various scenarios are simulating based on different access
the old FA that will be deleted8. methods: Distributed Coordination Function (DCF) and
Enhanced Distributed Coordination (EDCF)10, while
4. Performance Evaluation varying different network parameters: mobility speed,
and a number of users. Two types of traffic had been
The university has different departments scattered on considered: data (best effort), and video (Multimedia).
a different position, Figure 3 it plans to install a Wi-Fi The simulation result is given just for multimedia traffic
connection to cover the territory of its six departments being the most sensitive for network variation. Different
IP network planning is proving necessary. The choice is scenarios are generated by varying the number of clients
either to put all the departments in the same IP network (1,5, and 10) depending on mobility speed (1m/s, 5m/s,
(Level 2 mobility) or to put each department in a network 10m/s). In each Mobile IPv4 scenarios, a comparison
is separately or this case is mobility Level 3 (IP layer). The between access methods DCF and EDCF is done (the val-
purpose of this work is to plan the wireless network in ues used are by default).
such manner to ensure a continuity of service with bet-
ter QoS of mobile users even when moving from one
5. Results
network to another. The objective of this work is to give
a comparative study between level 2 - 3 mobility in this 5.1 Level 2 and Level 3 Mobility
experimental setting and to choose the appropriate access
Comparison
mode for the transfer of multimedia applications within
the university taking into account several parameters that Results give in terms of End to end delay that is defined
are mention in section 2. as the time to transmit a packet across a network from
source to destination. Figure 4-5 shows that in the case

Figure 3. University departments. Figure 4. Mobile IP packet end-to-end delay.

4 Vol 11 (14) | April 2018 | www.indjst.org Indian Journal of Science and Technology
 Hind Sounni, Najib Elkamoun and Fatima Lakrami

Figure 5. Level 2 mobility packet end-to-end delay. Figure 7. Level 2 mobility loss rate.

Figure 6. Mobile IP loss rate. Figure 8. Mobile IP throughput.

of level 2 mobility, better results are obtained than the by network congestion. It has measured the percentage
case when L3 mobility has occurred.In fact, when there of received packets compared to packets sent. As shown
is macro mobility, a triangular routing increases the delay in Figure 6-7 loss rate in L 2, mobility is lower than loss
of packet transmission.In both cases of mobility (MIPv4 rate obtained in the case of mobile IPv4. If the mobility
and mobility level 2), if the mobility speed increases the speed increases, the loss rate decreases in the case of
end to delay decreases. Packet loss occurs when one or the MIP and increases in the case of the mobility level
more packets of data traveling across a network fail to 2. Throughput represents the average number of bits
reach their destination. Packet loss is typically caused successfully received or transmitted by the receiver per

Vol 11 (14) | April 2018 | www.indjst.org Indian Journal of Science and Technology 5
 Performance Evaluation of Wireless 802.11n Using Level 2 and Level 3 Mobility

Figure 9. Level 2 mobility throughput. Figure 11. DCF vs EDCF loss rate.

unit time. Figure 8-9 shows a higher throughput for the case of 1m / s and 5m / s for different client numbers
mobility Level 2 than mobility Level 3, and it decreases in are almost the same. In the case of 10 m /s, the rate of
both cases if the speed of mobility increases. loss of EDCF is very low compare to DCF (Figure 11).
This comparison confirms the fact that EDCF gives better
5.2 DCF and EDCF Comparison in Mobile results compared to DCF in the case of level 2 and level
IPv4 3 mobility using MIPv4 for multimedia traffic, because it
The EDCF delay gives better results compared to the privileges the transmission of sensitive traffics.
DCF for the different scenarios that have been simulated
(Figure 10). The results of DCF and EDCF loss rate in 6. Conclusion
In this paper, an experiment comparison between differ-
ent types of mobility is given. The present study ensures a
continuity of services to mobile users with a better qual-
ity of service in Wi-Fi network. Therefore, a comparative
study through simulation has performed. Paper results
show clearly that deploying level 2 roaming give better
results compared with the deployment of level 3 roaming
using MIPv4, even whileincreasing the speed of mobil-
ity and the number of clients.In future work, a study and
an evaluation of mobility performance using mobile IPv6
will been done.

7. References
1. Gál Z, Karsai A, Orosz P. Evaluation of IPv6 services in
mobile WiFi environment, Computer Graphics. 2005; 231–
Figure 10. DCF vs EDCF packet end-to-end delay. 42.

6 Vol 11 (14) | April 2018 | www.indjst.org Indian Journal of Science and Technology
 Hind Sounni, Najib Elkamoun and Fatima Lakrami

2. Wi-Fi Roaming 101 - Cisco Blog. Accessed on: 05/06/2014. 7. Grenzinger Y, Jaquemet L. IP MOBILE OpNet. OPNET
Available at: https://blogs.cisco.com/wireless/wi-fi-roam- Technologies, Inc.; 2004. p. 1–9.
ing-101. 8. Min-hua Y, Yu L, Hui-min Z. The mobile ip handoff between
3. Hsieh R, Zhou ZG, Seneviratne A.S-MIP: A seamless hybrid networks. The 13th IEEE International Symposium
handoff architecture for mobile IP. In: INFOCOM 2003 on Personal, Indoor and Mobile Radio Communications;
Twenty-Second Annual Joint Conference of the IEEE 2002, 1. p.265–69.
Computer and Communications IEEE Societies, IEEE; 9. Lakrami F, Elkamoun N, Laouidya O. Performance
2003, 3. p. 1774–84. Crossref. comparison of wireless IEEE 802.11 a,b,g and n used
4. Park T, Dadej A. OPNET simulation modeling and for Ad-Hoc networks in an e-Learning classrooms net-
analysis of enhanced Mobile IP. 2003 IEEE Wireless work, International, Journal of Computer Science and
Communications and Networking, 2003. WCNC 2003; Information Security (IJCSIS). 2017 Sep; 15(9):229–35.
2003, 2. p. 1017–24. 10. Sounni H, Elkamoun N, Lakrami F. Performance evalu-
5. Talaat AH, Hussein NA, Elsayed HA, Elhennawy H. ation of 802.11 access methods with QoS enhancement
Real-time traffic performance for WiFi handovers over under mobility impact, International Journal of Computer
MIPv4 versus MIPv6, International Journal of Computer Applications. 2017 Dec; 179(2):36–41.
Applications. 2011 Jul; 26(8):10–15.
6. Yokota H, Idoue A, Hasegawa T, Kato T. Link layer assisted
mobile IP fast handoff method over wireless LAN networks.
In: ACM Press; 2002. p. 131–39. Crossref.

Vol 11 (14) | April 2018 | www.indjst.org Indian Journal of Science and Technology 7

View publication stats