Mini Thesis PDF

EVALUATING THE
PERFORMANCE OF MESH
NETWORK PROTOCOLS FOR
DISASTER SCENARIOS
by
Maggie Chimbwanda
A thesis submitted in partial fulfillment of

the requirements for the degree of
BSc. Honours Computer Science
University of the Western Cape
2011
Date: November 1, 2011

UNIVERSITY OF THE WESTERN CAPE
ABSTRACT
Evaluating the performance of Mesh network protocols in disaster scenarios
by Maggie Chimbwanda
Supervisor: Professor I.M. Venter

Co-Supervisor: Dr. W.D. Tucker
Department of Computer Science
Mesh networks are ad hoc networks where each node can supply connectivity to
adjacent nodes. Mesh networks originated in the military, but are now applied to
overcome some of the hurdles of traditional wireless deployments. Mesh clients
can either be stationary or mobile, and can form a client mesh network among
themselves with mesh routers. Mesh networks have routing protocols such as ad
hoc on demand distance vector routing, dynamic source routing, and optimized
link state routing.
In this project the performance of Mesh network routing protocols, when using a
user datagram protocol during disaster scenarios, will be investigated. These
protocols will be analyzed for efficient routing performance with respect to
different aspects such as network load, throughput, and delay.
A simulation tool, OPNET, will be used for modeling and simulating the
network to get the performance results of the protocols under the mentioned
aspects.
TABLE OF CONTENTS
Table of Contents .......................................................................................... i

List of figures .............................................................................................. iv
List of tables................................................................................................ vi
Acknowledgments ..................................................................................... vii
Glossary .................................................................................................... viii
Chapter 1 ...........................................................................................................1
User’s requirements document .......................................................................................... 1
Sketching the background .............................................................................................. 1
Performance of mesh networks .................................................................................... 3
Advantages of mesh networks....................................................................................... 4
Disadvantages of mesh networks .................................................................................. 4
User’s view of the problem ............................................................................................ 4
Brief description of users problems ............................................................................... 6
Project aim ...................................................................................................................... 6
What is expected from the testing of the performance of the protocols in
conjunction with mesh networks .................................................................................. 7
What is not expected from the testing of the performance of the protocols in
conjunction with mesh networks .................................................................................. 7
Conclusion ...................................................................................................................... 7
Chapter 2 ...........................................................................................................8
Overview of the protocols................................................................................................... 8
Introduction .................................................................................................................... 8
Mesh routing protocols and transport protocol ........................................................... 8
Why these protocols? .................................................................................................. 11
Why use the OPNET simulation tool? .......................................................................... 12
Conclusion .................................................................................................................... 13
Chapter 3 ......................................................................................................... 14
Simulation enviroment analysis ........................................................................................ 14
Introduction .................................................................................................................. 14
Brief description of complete simulation environment analysis ................................ 14
Simulation workflow .................................................................................................... 14
Brief description of a scenario ..................................................................................... 16
Simulation settings ....................................................................................................... 16
Building a network model for the scenario ................................................................. 18
Performance parameters ............................................................................................. 19
Coding description ........................................................................................................ 19
Simulation configuration .............................................................................................. 20
i
Conclusion .................................................................................................................... 21
Chapter 4 ......................................................................................................... 22
Prototype ........................................................................................................................... 22
Introduction .................................................................................................................. 22
Pilot study ..................................................................................................................... 22
Detailed description of the pilot study ........................................................................ 22
Pilot study results ......................................................................................................... 25
Conclusion .................................................................................................................... 29
Chapter 5 ......................................................................................................... 30
Simulation implementation approach .............................................................................. 30
Introduction .................................................................................................................. 30
Brief description of network simulation implementation .......................................... 30
Network modeling/design ........................................................................................... 32
Assigning values to the network components ............................................................ 34
Run simulation .............................................................................................................. 37
Conclusion .................................................................................................................... 37
Chapter 6 ......................................................................................................... 38
Simulation methodology ................................................................................................... 38
Introduction .................................................................................................................. 38
Simulation scenario (experiment)................................................................................ 38
Results ........................................................................................................................... 46
Challenges ..................................................................................................................... 49
Conclusion .................................................................................................................... 49
Chapter 7 ......................................................................................................... 50
Results analysis .................................................................................................................. 50
Introduction .................................................................................................................. 50
Results ........................................................................................................................... 50
Throughput ................................................................................................................... 50
Delay ............................................................................................................................. 56
Network load ................................................................................................................ 60
AODV in all experiments .............................................................................................. 66
DSR in all experiments .................................................................................................. 66
OLSR in all experiments ................................................................................................ 67
Comparative analysis of all experiments ..................................................................... 68
Conclusion .................................................................................................................... 72
Chapter 8 ......................................................................................................... 73
Final writeup ...................................................................................................................... 73
Introduction .................................................................................................................. 73
Project conclusion ........................................................................................................ 73
Future work .................................................................................................................. 74
Conclusion .................................................................................................................... 75
Appendices ................................................................................................ 76
Appendix A......................................................................................................................... 76
ii
Probes ........................................................................................................................... 76
Appendix B ......................................................................................................................... 78
Data from questionnaires ............................................................................................ 78
Appendix C ......................................................................................................................... 81
Term 1 Project plan ...................................................................................................... 81
Bibliography .............................................................................................. 89
Index .......................................................................................................... 92
iii
LIST OF FIGURES
Number Page
FIGURE 1: EXAMPLE OF MOBILE NODES IN A MESH NETWORK ................................................................. 2
FIGURE 2: UDP IN OSI MODEL ....................................................................................................... 11
FIGURE 3: OPNET WORKFLOW MODEL. ......................................................................................... 13
FIGURE 4: OPNET WORKFLOW ..................................................................................................... 15
FIGURE 5: AN EXAMPLE OF A SIMULATION ENVIRONMENT ................................................................... 20
FIGURE 6: PILOT EXPERIMENT SIMULATION WORKSPACE ..................................................................... 24
FIGURE 7: THROUGHPUT RESULTS IN ROUTING PROTOCOLS ................................................................. 26
FIGURE 8: DELAY RESULTS IN ROUTING PROTOCOLS ............................................................................ 27
FIGURE 9: NETWORK LOAD RESULTS IN ROUTING PROTOCOLS .............................................................. 28
FIGURE 10: FLOW OF THE PROJECT EXPERIMENTS................................................................................. 31
FIGURE 11: EXAMPLE OF A PEER TO PEER TOPOLOGY ............................................................................ 32
FIGURE 12: EXAMPLE OF A MESH NETWORK SETUP IN A DISASTER INCIDENT ............................................. 39
FIGURE 13: SCENARIO 1 WITH 10 NODES ........................................................................................... 42
FIGURE 16: DELAY RESULTS WITH 4 NODES………………………………………………………………………………….47
FIGURE 17: DELAY RESULTS WITH 7 NODES......................................................................................... 47
FIGURE 18: DELAY RESULTS WITH 10 NODES ...................................................................................... 48
FIGURE 19: DELAY RESULTS OF SCENARIO 1 ........................................................................................ 48
FIGURE 20: THROUGHPUT WITH 4 NODES .......................................................................................... 51
FIGURE 21: THROUGHPUT WITH 7 NODES .......................................................................................... 52
FIGURE 22: THROUGHPUT WITH 10 NODES ........................................................................................ 53
FIGURE 25: DELAY WITH 4 NODES ..................................................................................................... 56
FIGURE 26: DELAY WITH 7 NODES ..................................................................................................... 57
FIGURE 27: DELAY WITH 10 NODES ................................................................................................... 58
FIGURE 30: NETWORK LOAD WITH 4 NODES ....................................................................................... 61
FIGURE 31: NETWORK LOAD WITH 7 NODES ....................................................................................... 62
FIGURE 32: NETWORK LOAD WITH 10 NODES ..................................................................................... 63
FIGURE 35: ANALYSIS OF AODV IN ALL EXPERIMENTS .......................................................................... 66
FIGURE 36: ANALYSIS OF DSR IN ALL EXPERIMENTS.............................................................................. 67
FIGURE 37: ANALYSIS OF OLSR IN ALL EXPERIMENTS ............................................................................ 68
FIGURE 38: COMPARISON OF AODV, DSR, AND OLSR IN 4 NODES ....................................................... 69
FIGURE 39: COMPARISON OF AODV, DSR, AND OLSR IN 7 NODES ....................................................... 69
FIGURE 40: COMPARISON OF AODV, DSR, AND OLSR IN 10 NODES ..................................................... 70
iv
v
LIST OF TABLES
TABLE 1: OVERVIEW COMPARISON OF MESH ROUTING PROTOCOLS 9

TABLE 2: SIMULATION PARAMETER SETTINGS 18
TABLE 3: PILOT STUDY TESTING PARAMETERS 23
TABLE 4: NETWORK COMPONENTS 34
TABLE 5: DESCRIPTION OF ASSIGNING VALUES TO NETWORK COMPONENTS 36
TABLE 6: SCENARIO 1 NETWORK SIMULATION SETTINGS 41
TABLE 7: SCENARIO 2 CONFIGURATIONS 43
TABLE 8: SCENARIO 3 CONFIGURATIONS 45
TABLE 9: AVERAGE VALUES OF THROUGHPUT, DELAY, AND NETWORK LOAD FOR AODV, OLSR, AND DSR
FOR ALL EXPERIMENTS 72
vi
ACKNOWLEDGMENTS
I wish to express my sincere appreciation to Professor I.M. Venter and Dr. W.D.
Tucker for their assistance in the preparation of this project. In addition, special
thanks to University of the Western Cape (UWC) BANG Group members for
working together as a family. Thanks to the Center of Excellence (CoE) funding
of my project. Thanks also to the members of the school council for their
valuable input.
vii
GLOSSARY
Acknowledgment message (ACK). A message used to send and

acknowledgment from the destination to the sender that transmission is
successful in dynamic source routing.
Ad hoc on demand vector (AODV). A routing protocol that generates a route

request message to another node which has no route.
Dynamic source routing (DSR).A self-organizing and self-configuring protocol

that requires no existing network infrastructure or administration.
File transfer protocol (FTP). A standard network protocol used to copy a file
from one host to another over a TCP-based network, such as the Internet.
Hypertext transfer protocol (HTTP). A protocol that communicates between

web browsers with web servers and ensures secure communication.
Media access control (MAC). A unique identifier assigned to network

interfaces for communication on the physical network segment.
Mobile ad hoc network (MANET).A self-configuring network of mobile

devices connected by wireless links.
Multiple interface declaration (MID).A message that is sent to communicate

to other nodes in the optimized link state routing protocol.
Moving pictures experts group (MPEG-2 audio layer 3 / MP3). A common

patented digital audio encoding format for consumer audio storage as well
compression for the transfer and playback of music on digital audio players.
Network interface card (NIC). A computer component that connects a

computer to a computer network.
Network simulator version 2 (NS2). A discrete event simulator targeted at

networking research.
Optimized link state routing (OLSR). A routing protocol for mobile and ad
hoc networks, optimized to preserve bandwidth.
viii
Optimized network evaluation tool (OPNET). A simulation tool used to
model a network topology.
Personal digital assistant (PDA).A mobile device that functions as a personal

information manager.
Route request message (RREQ). A communication message that is sent in

the ad ho on-demand routing protocol to create a route from one node to
another.
Route reply message (RREP). A message that is sent after a request message
has been sent in the ad hoc on-demand routing protocol.
Request error message (RERR). A message that is sent if the destination

node cannot be reached by the source node in the ad hoc on-demand routing
protocol.
Short message service (SMS). The text communication service component of

phone, web or mobile communication systems.
Transmission control protocol (TCP). A protocol in the transport layer

responsible for transmitting data.
Topology Control (TC). A message that is sent in the optimized link state
routing protocol when updating the tables.
User datagram protocol (UDP). A protocol often used in videoconferencing

applications or computer games specially tuned for real-time performance.
Voice over internet protocol (VoIP).A transmission technology for delivery of

voice communications and multimedia sessions over internet protocols networks.
Worldwide interoperability for microwave access (WiMAX).A

telecommunications protocol that provides fixed and mobile internet access.
ix
Chapter 1
USER’S REQUIREMENTS DOCUMENT
Sketching the background

This project targets the establishment and management of communications
during disasters. Disaster events require real-time telecommunications which
can be only be realized if infrastructure with high speed performance is available.
Examples of disaster incidents where real-time communication would have been
useful but was lacking, include disasters such as hurricane Katrina that struck
New Orleans, United States of America (USA), in August 2005 and the
Indonesian tsunami that caused severe flooding and resulted in many deaths on
Boxing day (26thDecember) 2004. In both cases communication became a
problem because broadcasting towers were destroyed, and telephone lines were
washed away by the floods. The rescue service administration had to set up
emergency military public safety system networks to enable lifesaving
communication. These two scenarios would have been well served if the
construction of mesh networks were possible – with such networks it would
have been possible to have had effective real-time communication. But what
exactly are mesh networks?
Mesh networks are ad hoc networks originally used by the military. Each node of
this type of network supplies connectivity to adjacent nodes. The nodes can
either be fixed or mobile. Mobile nodes can be a mobile phone, a laptop, a
personal computer, a personal digital assistant (PDA), or an MP3player (See
Figure 1).
1
Figure 1: Example of mobile nodes in a mesh network
There are two types of mesh nodes: mesh routers and mesh clients. Mesh
routers provide bridge/gateway functionalities thus they are able to provide
access to other networks (e.g. internet, worldwide interoperability for microwave
access, etc.). Mesh routers act as the backbone for mesh clients. Mesh clients do
not have bridge/gateway functionalities when configured with wireless network
interface cards (NICs). Clients can perform peer-to-peer data transfers with
other clients when in ad hoc mode. Mesh clients can be computers or handsets.
2
Mesh networks have routing protocols that perform different tasks: ad hoc on
demand distance vector (AODV) routing protocols generate route request
(RREQ) messages if a source node wants to create a new route for a specific
destination. Dynamic source routing (DSR) are self-organizing and self-
configuring protocols that allow nodes to dynamically discover or source a route
for data packets from one router to another (known as hops) to any destination.
Optimized link state routing (OLSR) protocols are involved in updating and
maintaining information in a variety of routing tables.
The focus of this project is the evaluation of the above mentioned routing
protocols when using user datagram protocol (UDP) during disasters. UDP is
the protocol often used in videoconferencing applications especially tuned for
real-time communication (e.g. recording a real live disaster event in motion on
the web).
Performing an evaluation analysis on Mesh network protocols, could help with

the selection of the most reliable routing protocol. The network configuration
and its requirements for effective communication will also be considered.
Performance of mesh networks

Routing protocols play a major role in the performance of Mesh networks. This
encouraged various researchers to analyze the routing of protocols and
performance of the Mesh network under various network conditions: such as in
a grid environment (Usop, Abdullah, & Abidin, 2009) and in an ad hoc
environment (Rahman, Islam, & Talevski, 2010). Mesh networks under a grid
environment offers a form of network-distributed parallel processing system,
while under ad hoc environment are mobile and free to move around arbitrarily.
3
Some results showed that proactive protocols (OLSR) perform better than
reactive protocols (DSR, AODV). Reactive protocols are prone to buffer
overflow and packet drops over network layers (Ali & Ali, 2009).
Advantages of mesh networks

Mesh networks are easy to deploy, decentralized and need no existing
infrastructure. Quick deployment and minimal configuration of Mesh networks
makes them suitable to be used in emergencies such as disaster scenarios (Ali &
Ali, 2009).
Disadvantages of mesh networks

Mesh networks have volatile network topologies that make it hard to detect
malicious nodes. Security protocols that are available in wired network cannot
work in mesh networks (Kumar, 2009).
User’s view of the problem

The main motivation for this project is to find out how mesh network routing
protocols can best assist during disaster events. In order to do this, requirements
need to be gathered from users to identify the common problems they have
and/or will encounter that could be improved by the use of the mentioned
routing protocols. Two methods were used to gather user requirements: (i)
literature and (ii) interviews.
Literature
In disaster recovery scenarios reliable communication is of utmost importance.

With communication infrastructure destroyed or areas where there is little
communication infrastructure, disaster response teams need to setup their own
communication infrastructure. Mesh networks are ideal in such situations since it
4
can deploy available devices such as mobile phones, laptops, MP3 players, etc. as
part of the mesh network.
According to the article “Tactical Communications vol. 3 issue 12”, factors that
make communication reliable are that users should be able to communicate in
real time and transfer information between each other. Mesh network should be
scalable to support the increased number of users when browsing the web or
transferring information. Mesh networks need to have components such as
broadcasting fax, websites (for emails and sending/retrieving information),
radios, and conference calls (Preparedness, 2008).
According to Townsend during the rescue period after the hurricane Katrina,
most of the rescue teams and victims had to communicate more using mobile
phones to call each other. Some of the victims could and some could not make
calls or send SMS to their friends and family to inform them of their locations. It
states that voice over IP is said to be the way to engineer survivability during
disaster scenarios (Townsend & Moss, 2005).
Two communication difficulties have been identified voice and video

communications limitations. The efficient performance of communication
infrastructure is paramount.
Interviews
Two persons were interviewed using a set of questions (see Appendix B).Both
interviewees have never experienced a disaster event; however, the first
interviewee was aware of the applications and devices used to assist with the
deployment of a mesh network. The second interviewee had knowledge of voice
broadcasting applications such as SMSs if a disaster were to occur. From the two
interviews it can be concluded due to the lack of disaster experience and lack of
experience of not constructing a mesh network, it likely that the two personnel’s
5
are not aware of some the aspects that need to be considered for construction
mesh networks, such as the routing protocols needed to be considered to make
communication effective in an event of disaster.
Brief description of users problems

The main problems identified in the literature reviews from the users were
mainly;
Voice communication:
Users want to be able to communicate using voice communication because it is
fast and immediate, and it is a communication mechanism that is used every day
and every minute.
Video communication:
Users want to be able to communicate using video communication to talk to
other rescuers in real time. This is because video communication can provide
both voice and video feeds of things happening instantly.
Project aim
The main aspect of this project is to evaluate which of the routing protocols will
provide better performance during disasters by reducing the problems
experienced with voice communication, and video communication. This is
because first responders to a disaster event need a communication that is flexible
and easily deployable to be able to communicate with other rescue teams by
means of voice and video. They also want to communicate with live video calls
and/or video conferencing to talk or show live feeds of the current situation in
their environment. Mesh networks are sensitive to throughput, delay and traffic
load; hence these aspects will be evaluated. The network load, number of nodes,
6
and speed of the nodes play an important role in efficient results of traffic
routing for protocol performance.
What is expected from the testing of the performance of the protocols in

conjunction with mesh networks
The performance of the different Mesh network routing protocols (when using
UDP) will be evaluated. An analysis of the performance of the different routing
protocols will be given and in order to find best protocols for use during
disasters for efficient communication. The routing protocol that has the lowest
performance in the metrics considered will be the best protocol to use.
What is not expected from the testing of the performance of the protocols
in conjunction with mesh networks
The optimized network evaluation simulation tool (OPNET tool) will be used
for the evaluation of the routing protocols. The evaluation will not be an
exhaustive evaluation. Only the routing protocols will be tested in conjunction
with UDP. The results will thus only reflect these conditions.
Conclusion
In this chapter, the use of mesh networks in disaster situations was discussed.
The most important requirements for such a network were identified namely:
voice broadcasting, and video calls/conferencing. In this project these aspects
will be addressed by testing of a simulated network using a simulation tool
OPNET. The next chapter focuses on an overview of the protocols that will be
considered in this project in order to simulate and test the protocol
performances for these aspects.
7
Chapter 2
OVERVIEW OF THE PROTOCOLS
Introduction
In the previous chapter, we looked at the background of mesh networks, their
advantages, the project motivation, user’s view of the problem, and what is/not
expected from the performance of the protocols. Based on the information
gathered, the main problem is identifying the most reliable, efficient and accurate
routing protocol that will be suitable for Mesh networks in disaster scenarios. In
this chapter an overview of the protocols to be considered in this project will be
looked into detail individually.
Mesh routing protocols and transport protocol

Mesh network routing protocols AODV, DSR, and OLSR can be categorized as
being; reactive or proactive. Reactive protocols (such as AODV and DSR) are
protocols that are called on demand (i.e. when they are needed) and routes are
built by sending route requests through the network. Proactive routing protocols
(such as OLSR) are protocols that maintain and build routing information by
periodically transmitting the control messages. Transport protocol UDP is
categorized as a connectionless-oriented protocol, meaning that transmitting of
data does not require the source to verify the readiness of the receiver. Overview
comparison of the mesh routing protocols is shown in Table 1 below.
8
Type of Name of Characteristics
Protocols Protocols
Routing Type of Frequency of Worst case Multiple

Philosophy Routing updates routes
Routing DSR Reactive Source As needed Full flooding Yes
AODV Reactive Hop-by-hop As needed Full flooding No
OLSR Proactive Hop-by-hop Periodically Pure link state No
Transport UDP Connectionless Host-to-host Periodically Congestion No
Table 1: Overview comparison of mesh routing protocols
Ad hoc on-demand distance vector routing protocol (AODV)

AODV is a reactive routing protocol that generates a route request (RREQ)
message to another node which has no route. This protocol has three types of
control messages for route maintenance; route request (RREQ), route reply
(RREP), and route error (RERR) (Ravikumer & Chittamuru, 2010).
Dynamic source routing protocol (DSR)

DSR is a self-organizing and self-configuring protocol that requires no existing
network infrastructure or administration. It allows nodes to dynamically discover
a source route across multiple hops to any destination. DSR comprise of two
mechanisms: route discovery and route maintenance (Ravikumer & Chittamuru,
2010). DSR also has the same routing messages as AODV, but DSR uses RREQ
and RREP messages for route discovery and RERR and ACK
(acknowledgement) messages for route maintenance. Acknowledgement
9
messages are sent when a destination node successfully receives packets that are
transmitted from the sender node in response to notify that transmission is
successful.
Optimized links state routing protocol (OLSR)

OLSR is a proactive routing protocol involved in updating and maintaining
information in a variety of tables. It is a table-driven protocol used to preserve
bandwidth. The protocol uses the link state algorithm throughout the network.
OLSR has three basic types of control messages, hello, topology control (TC),
and multiple interface declaration (MID) (Ravikumer & Chittamuru, 2010). Hello
messages are sent in a predetermined manner between nodes. If nodes are
neighbors and a hello message is sent from node A to B, with B being the
receiving node, then an asymmetric link is made. This is also applied if the
situation was vice-versa (B sending to A). But if a two-way communication is
made, this is known as a symmetric link and TC messages are broadcasted.
User datagram protocol (UDP)

UDP is a simple OSI transport layer protocol for client/server network
applications based on internet protocol (IP), see Figure 2. UDP is the main
alternative to the transmission control protocol (TCP), one of the oldest network
protocols in existence, introduced in 1980 that is responsible for transmitting
data, because UDP is compatible with packet broadcasting and multicasting and
also reduces latency compared to TCP. UDP differs from TCP in the sense that
UDP is a connectionless protocol that does not require verifying the readiness of
the receiver before transmitting information. On the other hand, TCP is a
connection-oriented protocol that requires the handshake mechanism to set up a
connection (Kurose & Rose, 2010).
10
Figure 2: UDP in OSI model
UDP is often used in videoconferencing applications or computer games

specially tuned for real-time performance. To achieve this higher performance,
the protocol allows individual packets to be dropped (with no retries) and
packets to be received in a different order than they were sent as dictated by the
application. The dropped packets only results with slight degradation of quality
of video streaming rather than large delays if lost packets were to be
transmitted(Kurose & Rose, 2010).
Why these protocols?

In mobile ad hoc networks (MANET), nodes are not aware of the network
topology, so the routing protocols AODV, DSR, and OLSR will discover the
topology by receiving the broadcast message from neighboring nodes and need
to respond to the message accordingly, such as when routes are being established
or when an update is being made. Comparison of these routing protocols in
conjunction with the UDP protocol, when considering the metrics throughput,
11
delay, and network load, will determine which routing protocol will perform best
in ad hoc situations.
Why use the OPNET simulation tool?

The optimized network evaluation tool (OPNET) is not open source software,
but licensed software that requires application for the license to have it installed
on a personal computer. The license was readily available for this project, and by
using the OPNET simulation tool it will be possible to create a virtual network
to test and compare routing protocols in conjunction with UDP. OPNET allows
the specification of propagation models depending on the type of network
constructed. The tool comprises of: a graphical interface of predefined nodes;
well defined documentation and the necessary algorithms that mimic the
protocols that will be tested, thus OPNET will be the tool of choice for this
project to test and evaluate these routing protocols. The OPNET workflow
allows four activities: model design; collecting statistics; running simulation; view
of results and analysis (See Figure 3).
12
Model Design
View and Analyze Collect Statistics

results (protocols, testing
metrics, etc.)
Running
Simulation
Figure 3: OPNET Workflow Model.
Conclusion
In this chapter an overview of the mesh network routing protocols with the user
datagram protocol have been discussed in order to get a clear understanding of
why these protocols are important in the implementation of this project. In the
following chapter the design aspects of testing the simulation environment will
be analyzed.
13
Chapter 3
SIMULATION ENVIROMENT ANALYSIS
Introduction
In the previous chapter the mesh network routing protocols were individually
introduced. In this chapter the project simulation environment analysis plan will
be discussed, it will include a brief description of the planned analysis, the
simulation environment, a scenario description, description of how a network
for a specific scenario will be modeled, performance metrics, code generation
description, simulation settings, and what the design will look like.
Brief description of complete simulation environment analysis

The simulation environment using the simulation tool OPNET will be to test
and evaluate the performance of routing protocols with the aim of finding the
best routing protocol for use during a disaster.
This simulation design will be used with different scenarios to determine the
optimal protocol. The requirements of the different scenarios include the
following: type and number of devices users might use; the geographical radius;
protocols to be assigned to a network; and different testing parameters. Based on
these requirements from the scenario, five simulations will be conducted based
on the parameters mentioned in the first chapter (throughput, network load, and
delay).
Simulation workflow
As mentioned in the previous chapter, OPNET (version 16.1) was chosen as the
simulation tool of choice for this project because the license for this software
was readily available to be used in this project. Application for the license is
required to enable installation on a personal computer so that modeling of a
14
network and evaluation of the protocols behavior in the different scenarios can
be implemented. It has a flexible graphical interface to simulate and view results.
The basic workflow of the OPNET model (refer to Figure 4), shows the
necessary procedures that will be followed to conduct this project.
Figure 4: OPNET Workflow
The model design stage involves selecting and dragging the necessary objects
such as: nodes, servers, routers, etc. onto the simulation workspace. This stage
also involves the selection of routing protocols to be used in the network,
applications to be applied to a scenario (e.g. voice calling), mobility speed, and
simulation running time. Collection of statistics stage is where statistics are
collected based on two options: object statistic (from individual nodes in the
network) or global statistics (from the entire network). These statistics include
the routing protocols, testing parameters, and the application assigned in the first
15
stage. Run simulation is the running of the modeled network based on the time
set in the first stage. The viewing and analyzing of the results in OPNET is based
on two metrics (i.e. global or object) this gives the option to compare the routing
protocol performance whether of individual nodes or the network as a whole.
Brief description of a scenario

The project will be to test protocols in terms of specific scenarios, for example;
“Cape Town, South Africa suffers a small earthquake disaster. It causes all internet
connections in the area to malfunction due to collapsed buildings and communication towers.
First responders at the scene consist of 4 rescuers. For easy movement around the scene the rescue
teams have a mobile phone, and two handheld radios that can be used for communication. A
stationery point (tent) is setup for the mobile rescuers to report on the evolving events around
them to communicate with each other, other approaching rescuers and the victims’ families”.
They need to set up a wireless ad hoc network with the 4 communication

devices they have. But for them to know that communication will be efficient
and reliable they need to know which routing protocols would be best to be
used. The rescue team may grow in size depending on how bad the disaster
event is. Thus the number of devices will increase to make communication much
more efficient. A model of this scenario will be modeled using the OPNET
simulation tool. Several simulation tests will be conducted with different number
of nodes to determine which routing protocol performance gives the best results
under the testing parameters throughput, network load, and delay.
Simulation settings
The performance of the routing protocols (AODV, DSR, and OLSR) with UDP
will be tested using a simple simulation setup. Most of the variables are static; the
only dynamic parameters will be the number of nodes. The performance metrics
will be tested in each simulation experiment so that comparison of the routing
16
protocols can be effective. Based on the scenario described above, the following
simulation settings (refer to Table 2) will be used to evaluate the performance of
the routing protocols.
17
Experiment Parameters
No. of Routing Other Performance Simulation Mobility rate Simulation

nodes protocols protocols metrics radius time
1 4 AODV UDP Throughput, 2000m x 2000m 10meters/sec 10min

DSR delay, network
OLSR load

DSR delay, network
OLSR load

DSR delay, network
OLSR load

DSR delay, network
OLSR load

DSR delay, network
OLSR load
Table 2: Simulation parameter settings
Building a network model for the scenario

Creating a network model with the simulation tool OPNET will be conducted
by first creating a blank scenario from a menu (not retrieved from a saved file).
Insertion of nodes and network devices can be done automatically by choosing
from a wizard menu, or manually dragging components to be used from an
object palette on to the created blank scenario. The nodes will be configured by
assigning the routing protocols and parameters to be tested. The mesh routing
18
protocols (AODV, OLSR, and DSR) in conjunction with UDP will be evaluated
based on the parameters throughput, delay and network load.
Performance parameters
The performance metrics: network load; throughput; and delay will give different
results of the different simulations.
Network load:
Network load will test the amount of data traffic carried by the network. Each
routing protocol will be tested based on the weight of the network (i.e. number
of nodes in the network) and the volume of the data received between nodes
during simulation.
Throughput:
Throughput will test the amount of data that reaches the receiver from the
source to the time taken by the receiver to receive the last packet. The routing
protocols will be tested on throughput when using UDP between nodes (i.e. if
node is making a voice communication).
Delay:
Delay will test the number of time taken by packets to pass through the network.
This parameter will test the routing protocols acceptance to the various
constraints when the UDP protocol is used in conjunction, as voice applications
are very sensitive to delay.
Coding description
The OPNET simulation tool allows code construction to be done in a high-level
programming language such as C or C++. In this project, C programming
language will be used for the code construction. The coding documentation will
be discussed further in Chapter 6.
19
Simulation configuration
The project’s proposed network simulation configuration will test three different
scenarios with each scenario having an increase in the number of nodes to test
these routing protocols. An example of how the modeled network will be
configured with 10 nodes for the routing protocols OLSR, AODV, DSR and the
UDP protocol is shown in Figure 5. Additional devices include an, application
configuration, and profile configuration. The testing parameters throughput,
delay, and network load are provided here.
Figure 5: An example of a simulation environment
20
Conclusion
In this chapter a brief description of the simulation environment analysis plan
was discussed, such as the environment in detail, the scenario detail, modeling of
network for the scenario, performance metrics, simulation settings, code
generation description and what the simulation environment will look. In the
chapter following, a prototype of the simulation environment analysis will be
developed.
21
Chapter 4
PROTOTYPE
Introduction
The previous chapter gave a detailed description of the simulation environment,
and what aspects will be considered during testing. In this chapter the setting up,
testing and the results of a pilot study will be discussed for the prototype.
Pilot study
This project’s prototype is based on a pilot study. A pilot study also known as a
pilot experiment is a small scale preliminary study conducted before a large-scale
quantitative project is implemented, in order to check the feasibility or to
improve the design of the whole project (Bille.R, 2010). A pilot study is used to
test the design of the full-scale experiment which can be adjusted (Haralambos,
2000).
This project is more of testing and evaluation of mesh routing protocol, thus
using a pilot experiment will best demonstrate the design of the full-scale
experiment.
Detailed description of the pilot study

The pilot study will give a brief overview demonstration of how the simulation
environment analyzed in the previous chapter will be tested for the whole
project. The pilot experiment will follow the OPNET simulation work flow
discussed in the previous chapter (see Figure 4).
This pilot experiment will be based on an example scenario of a small disaster

event that will not require a lot of rescue members, thus only four nodes will be
used for this study. The network model will be an example of a peer-to-peer
22
network because peer-to-peer networks are typically used for ad hoc networks
such as mesh networks, thus no need for a server. The settings for the pilot
experiment will include the routing protocols (AODV, DSR, and OLSR) with
the transport protocol UDP, (See Table 3), of which all three routing protocols
will be tested based on the three testing parameters: throughput, delay, and
network load. In comparison to the table in the previous chapter, the pilot setup
will be different in the sense that the number of nodes and number of routing
protocols to test will be dynamic while the rest of the parameters will be static.
Experiment Parameters
Name
No. of Routing Other Performance Simulation Mobility rate Simulation

nodes protocols protocols metrics radius time
Pilot 4 AODV, UDP Throughput, 100m x 100m 5 meters/sec 10min

OLSR,DSR delay, network
load
Table 3: Pilot study testing parameters
Modeling of the network involves using the OPNET wizard menu to first create
a blank workspace and then dragging the nodes and other network devices from
an object palette onto the workspace. Each node is then assigned the routing
protocols (AODV, DSR, and OLSR) and transport protocol UDP. Other
parameters assigned are applications such as video conferencing because this is
one of the user requirements identified in Chapter 1.
After these parameters are entered into the OPNET network model, the pilot
experiment will run and test the performance of the routing protocols, and based
23
on the given results a comparison of the routing protocol will be analyzed to give
the best performance under the testing parameters (throughput, delay, and
network load) of routing protocol performs best. The simulation tool OPNET
shows results based on two aspects; global (as a whole network) or object (as
individual nodes). Retrieving results by object is exhaustive hence global results
will be used to compare the performance of the pilot experiment. The Figure
6below shows the model of the pilot experiment simulation workspace.
Figure 6: Pilot experiment simulation workspace
The above figure shows the network model for the pilot experiment. The red
line represents an outline or boundary or domain of a disaster scenario, and the
24
four nodes inside the red line represent the mesh network construction within
the domain. The four nodes are mobile within the domain thus if a node is
configured for mobility it is represented by a white arrow on each node (see
Figure 6). The two devices outside the domain are the profile definition node
and the application definition node. The application definition defines the
application defined to the nodes in the network; in this case a video conferencing
application is assigned to this experiment according to one of the user
requirements. The profile definition defines the activity patterns of a node in
terms of the applications used; in this case the profile described for this
experiment is multimedia expert.
Pilot study results

Figure 7 shows the pilot study results of the performance metric throughput, and
in Figure 8 shows the results of delay, whilst Figure 9 shows the results of
network load in all three routing protocols (AODV, DSR, and OLSR) when
using UDP. The graphs are color coded for easier comparison between the
protocols. As shown in Figure 7, Figure 8, and Figure 9, AODV protocol is
represented by a blue line, DSR by a red line and OLSR by a green line. The
results show the global results only. As mentioned before, global results show
results gathered from the entire network, and object results are gathered from
individual nodes.
25
Figure 7: Throughput results in routing protocols
26
Figure 8: Delay results in routing protocols
27
Figure 9: Network load results in routing protocols
For global results, it shows that DSR and OLSR have high throughput and a
constant speed when sending packets compared to AODV which resulted in low
throughput and a loss of packets when the simulation time was at 4mins. For
delay results, AODV showed the highest delay and again had a loss of packets at
4mins of simulation time; OLSR having the second highest delay compared to
DSR that had low delay. Both OLSR and DSR did not experience packet loss
during the simulation. Network load results for DSR and OLSR show high load
28
and constant sending of packets compared to AODV network load which
showed low results and again packet loss at 4mins of simulation time.
Based on the pilot experiment results it may be concluded that DSR is the best
routing protocol to be used in disaster scenarios when constructing mesh
networks for efficient communication.
For the next stage of the project, project implementation, the full-scale
experiment will be conducted and these (pilot) results may tend to change as the
number of nodes will increase to give different results. Also during
implementation it will be taken into consideration why the routing protocol
AODV experienced packet loss for all three performance metrics at the same
simulation time of 4mins.
Conclusion
The chapter gave an overview of how the prototype pilot study settings were
conducted and gave an insight of how the whole project will be implemented.
The pilot study gave comparison results of the routing protocols. The next
chapter will discuss the implementation of the full scale experiment with some
knowledge of how the pilot experiment was performed.
29
Chapter 5
SIMULATION IMPLEMENTATION APPROACH
Introduction
The previous chapter discussed the pilot experiment which gave a brief
description of how the implementation of the overall project will be conducted.
This chapter will now explain the implementation planning of the full-scale
project in much more detail, explaining all the configurations that are required to
conduct implementation.
Brief description of network simulation implementation

The project implementation will follow an incremental development whereby
modeling each of the six experiments will be continuous and modified (Boehm,
1986). Five simulated network models based on five experiments was
implemented (Table 2). In each experiment most of the parameters were static
(only the permutation of types of nodes was dynamic) (see Table 6). The number
of nodes was changed and increased for each individual experiment, this is
because in Chapter 1 it is mentioned that the motivation of this project is to find
the best routing protocol to be used in disaster events, so the increase of the
number of nodes will help to analyze the performance of routing protocols when
nodes grow in size if a bigger disaster were to occur.
30
40
20 nodes
10 nodes
nodes
7
nodes
4
nodes
Figure 10: Flow of the project experiments
For the first experiment 4 nodes were used to simulate the network and the
performance metrics: network load; throughput; and delay were used to test the
routing protocols AODV, OLSR, and DSR. These parameters were also
considered for the other four experiments; experiment 2 with 7 nodes,
experiment 3 with 10 nodes, experiment 4 with 20 nodes, and experiment 5 with
40 nodes (see Table 2).
The modeling of the network was done using the simulation tool OPNET
(version 16.1). The modeling process followed the OPNET workflow model
(see Figure 4) a modeling design stage; an assign values stage; and a running
simulation stage was executed and is discussed in this chapter of the project,
whilst testing, viewing and analyzing of the results will be discussed in the
chapter following the methodology.
31
Network modeling/design
The first step of the OPNET workflow modeler is modeling or the design of the
simulation network.
The network topology modeled is a peer-to-peer network (same as in the pilot

experiment) because Arnold et al state in a report that peer-to-peer network
architectures offer a promising solution to the challenges of information-sharing
in disaster events (Arnold, et al., 2004). Thus this kind of network topology is
beneficial in mesh networks for disaster events because they are easy to deploy
and every node acts as a server at the same time as a client, thus no need for a
router or a server (see Figure 11).
Figure 11: Example of a peer to peer topology
32
Implementation of the modeling of the network starts by first generating a blank
scenario/workspace with the startup wizard. OPNET allows manually modeling
of the network (entering parameters that satisfy the user) or automatically
modeling (selecting from a pre-defined list from the wizard). In this case the
modeling was done using the automatic option because it is a less exhaustive
process compared to the manual configuration. With manual configuration each
of the network components are dragged from the object palette onto the
workspace, one-by-one. The components that will be included in this project
include; nodes, application configuration, profile definition, and network
domain.
33
Table 4: Network components
Assigning values to the network components

The values were assigned in two steps; first each individual (object) value was
assigned to each node and it included; routing and transport protocols, profiles,
applications, mobility speed, and IP addresses. Secondly values were assigned
34
globally to the whole network, it included; simulation time, mobility rate, and
assigning results parameters.
Assigning of these values were based on the requirements gathered in Chapter 1,

thus the routing protocols that were assigned were AODV, DSR, and OLSR; the
transport protocol UDP; and the applications that were assigned were voice
over IP as identified in the user requirements. The testing parameters that were
assigned globally (to the whole network), were: delay; throughput; and network
load. These made it possible to analyze the performance of the routing protocols
chosen for the three scenario experiments (refer to Table 5).
35
Table 5: Description of assigning values to network components
36
Run simulation
The last stage of the implementation of the project experiment is running the
simulated network to test the routing protocols. All the nodes and entities in the
network need to be assigned and configured before “running”. In the menu bar
selecting DES>Run Discrete Event Simulation runs the simulation based on the
simulation time set during the assigning stage.
Conclusion
In this chapter an overview of the project implementation configurations steps
have been discussed. The following chapter explains the testing of the three
scenarios which consist of the five experiments.
37
Chapter 6
SIMULATION METHODOLOGY
Introduction
In the previous chapter the simulation implementation configurations, the
parameter settings, entities, and values that make up the network for the
experiments were discussed. In this chapter explains the implementation of the
three experiments and its associated challenges are presented.
Simulation scenario (experiment)

In Chapter 3 (p16), a brief description of a scenario was discussed. The scenario
describes the occurrence of an earthquake in an area in Cape Town, South Africa
where the communication towers and lines are down hence causing
communication impossible. A small number of first responders (four) arrive at
the scene to set up an emergency temporary network for rescue teams to
communicate and help victims. Whatever the case, rescuers need a flexible, easily
deployable, and quick establishment of communication infrastructure that is not
fixed (wired), thus mesh networks are effective for this event. An example of
disaster scenario with a mesh network setup is shown in Figure 12.
38
Figure 12: Example of a Mesh network setup in a disaster incident
A stationary point (tent) is set up where rescuers can report to other rescuers and
family members as well assist in medical assistance. Other team members are
mobile using devices such as mobile phones, PC tablets, smartphones, handheld
radios, etc. to communicate with each other whilst finding victims or reporting
on the event.
The number of rescue team members will increase if it is discovered that the
disaster effects are much worse and more rescuers are needed at the site. Before
the network is setup, a simulation test is performed to test the performance of
routing protocols that are reliable and efficient for communication. Hence three
different network scenarios that consist of five experiments are simulated and
39
tested, each representing the scalability of the network when the number of
rescue members coming onto the site increases.
Each scenario comprised the routing protocols AODV, DSR, and OLSR; and
the transport protocol UDP. A running time of 10 minutes was applied to all
five experiments at a mobility rate of 10 meters/sec within a radius of 2000m x
2000m. For each scenario the behaviour of the routing protocols was considered
in terms of the performance metrics; throughput, delay, and network load. The
test results for all five experiments were global results (from the network as a
whole) and not object results (from individual nodes). Other aspects that were
considered when analyzing the results were the scalability and how the running
time is affected depending on the scalability of the network.
Scenario 1:
For the first scenario three experiments were conducted, with a difference in
number of nodes, area size, and node permutations (Table 6). The main aim of
conducting the first two experiments (4 nodes and 7 nodes) was to see if
different permutations and scalability of nodes would give a difference in results,
and based on the results, the third experiment was implemented (10 nodes)
which is the main project simulation requirements (Chapter 3 p18).
The three different permutations shown below in the table were the ones that
showed a change in results. Table 6 gives the details of the experiments (scenario
1).
40
Table 6: Scenario 1 network simulation settings
The metrics throughput, delay and network load were used to evaluate whether
the routing protocols’ (AODV, DSR, and OLSR) performance are reliable and
efficient during communication. As mentioned in Chapter 1 (p6) one of the
user’s requirements identified was to be able to communicate by voice, so the
application for this scenario was voice-calling; and the transport protocol for this
application is UDP.
The simulation radius describes the geographical location size of the network
setup where the rescuers can be mobile at a rate of 10 meters/sec within the
41
specified radius. The packet size describes the size of the voice packet that was
sent from one node to the other. Simulation radius, mobility rate, and packet size
helped in analyzing the performance of the routing protocols. The time it takes
to run the simulation (real-time not simulation) analyze any errors that might
occur (e.g. packet loss during transmission).
Simulation time is the time set to run the simulation. For this project and
scenario the run time is set to 10 minutes. The simulation time is the time
recognized in the OPNET modeler and not the actual wall-clock time. Figure 14
shows a simulation setup for scenario 1.
Figure 13: Scenario 1 with 10 nodes
42
Scenario 2:
Based on the difference of results retrieved from the experiments conducted in
scenario 1, it was safe to test the project simulation by increasing the nodes.
Scenario 2 consisted of 20 nodes and a radius size of 2000m x 2000m, a
simulation time of 10 min at 10metres/sec mobility rate, using voice application.
Because permutation of 20 nodes was exhaustive, this experiment only tested
nodes that included 5 mobile phones, 5 radios, 5 laptops, and 5 smart phones.
Table 7 gives the details of the experiment (Scenario 2). Figure 14 shows the
simulation setup for Scenario 2.
Table 7: Scenario 2 configurations
43
Scenario 3:
This scenario simulates a number of 40 nodes with a radius size of 2000m x
2000m, a simulation time of 10 min at 10metres/sec mobility rate, using voice
application. Permutation of 40 nodes was exhaustive so this experiment only
tested nodes that included 10 mobile phones, 10 radios, 10 laptops, and 10 smart
phones. Table 8 gives the details of the experiment (Scenario 3). Figure 15 shows
the simulation setup for Scenario 2.
44
Table 8: Scenario 3 configurations
45
Results
The results of the above mentioned experiments are shown in Figure 16, Figure
17, & Figure 18. These results are from Scenario 1 and only show the testing
parameter delay from the following selected permutations: 2 cellphones and 2
laptops (experiment 1); 3 cellphones, 2 handheld radios and 2 laptops
(experiment 2); and 4 cellphones, 4 handheld radios and 2 laptops (experiment
3). The rest of the results for the other test metrics (throughput and network) for
Scenario 1 and 2 will be shown and analyzed in the following chapter because
this chapter only discusses the project implementation.
46
Figure 16: Delay results with 4 nodes Figure 17: Delay results with 7 nodes
47
Figure 18: Delay results with 10 nodes
In Figure 19 the scalability of the routing protocols is illustrated and compared.
Delay results of AODV, DSR, OLSR with 4, 7, and

10 nodes
0.4
0.35
0.3 AODV_4
DSR_4
0.25
Delay in seconds
OLSR_4
0.2 AODV_7
DSR_7
0.15
OLSR_7
0.1 AODV_10
DSR_10
0.05
OLSR_10
0
0 2 4 6 8 10 12
-0.05
Simulation time in minutes
Figure 19: Delay results of Scenario 1
From this figure it can be seen that when the number of nodes are increased, the
delay increases. The different routing protocols perform differently.
48
Challenges
With experiment 1 in Scenario 1 (4 nodes), the DSR routing protocol was able to
simulate and produce efficient results for both voice and video applications.
When the numbers of nodes were increased (Experiment 2 and 3 with 7 and 10
nodes) the simulation failed for video-conferencing. This occurred because DSR
routing processes in the nodes failed to recognize and set the size of the packets,
despite reducing the packet size. The voice application was successful, because it
was independent of the number of nodes or area size, all three routing protocols
simulated problem free.
Conclusion
In this chapter the project implementation, testing of the three scenarios and
their experiments were discussed. A discussion of the results of the overall
project will be discussed in the next chapter so that the best routing protocol can
be predicted based on the analysis of the results.
49
Chapter 7
RESULTS ANALYSIS
Introduction
In the previous chapter the simulation, testing, and the implementation of the
routing protocols (when using a voice-calling application) were discussed. In this
chapter the results of all the five experiments are discussed and analyzed to give
the results of the best performing mesh routing protocol.
Results
The test results of the routing protocols AODV, DSR, OLSR were obtained in a
graphical format because OPNET results are presented and compared using
graphs. Each of the routing protocols is represented in a color-coded graphical
line: OLSR protocol is represented by a green line; DSR by a red line; and
AODV by a blue line. As mentioned in Chapter 1, these routing protocols are
analyzed based on the performance metrics: throughput; delay; and network
load. Five experiments were executed using the OPNET modeler. A comparison
was done to give better view of how each of these protocols performed when
the number of nodes were increased.
Throughput
Throughput results of all five experiments are shown below individually.
Throughput is the total amount of data that reaches the receiver (from the
source) compared to the time taken by the receiver to receive the last packet.
Throughput results increased for each experiment when the number of nodes
increased thus giving a high performance for the different routing protocols.
50
Figure 20: Throughput with 4 nodes
51
52
53
54
From these results it can be seen that DSR has the highest throughput (Figure
20) when the number of nodes are less, but when the nodes increase, OLSR had
a high throughput (see experiment 2 and 3 (7 and 10 nodes)). AODV had a
much higher throughput when there are more than 20 nodes, thus AODV
comparatively performs better than other routing protocols (DSR, OLSR) when
using the transport protocol UDP for a voice traffic application for the highest
number of nodes.
55
Delay
Delay is the time taken by packets to pass through the network. Figures 25-29
show the delay results of all five experiments as individual graphs. Delay results
increased when the number of nodes for each experiment increased.
Figure 25: Delay with 4 nodes
56
57
58
59
From these results it can be seen that AODV has high network load when the
number of nodes are less (experiment 1 and 2), but when the nodes increase to
10, 20, and 40 nodes, DSR has the highest delay. Thus DSR out-performs
AODV and OLSR when using the transport protocol UDP for a voice traffic
application.
Network load
Network load is the amount of data traffic carried by the network. Figures 30-34
show network load results of all five experiments. The network load experiment
60
results also show an increase in performance when the number of nodes
increased.
Figure 30: Network load with 4 nodes
61
62
63
64
From these results it can be concluded that DSR presents with the highest
network load for experiments with 4, 7, 10, and 20 nodes respectively. Whilst
AODV shows the highest network load for 40 nodes. Based on the first four
experiments DSR outperforms AODV and OLSR (even though AODV
outperforms DSR when 40 nodes are deployed). Further testing (with more
nodes) is required to determine if AODV outperforms DSR.
65
AODV in all experiments
Routing protocol AODV was simulated in all five experiments and results are
retrieved from the testing parameters throughput, delay, and network load. The
graph shows that AODV (when using UDP) has much high throughput in 40
nodes compared to when nodes were 10 and less. The behaviour of AODV is
almost the same for network load and delay for 4, 7, 10, 20, and 40 (i.e. results
being in the same scale range).
ADOV results for all experiments

16000000
4_Delay (sec)
14000000
7_Delay (sec)
12000000 10_Delay (sec)
20_Delay (sec)
10000000
40_Delay (sec)
8000000 4_Throughput (bits/sec)
7_Throughput (bits/sec)
6000000

4_Networkload (bits/sec)
0
0 2 4 6 8 10 12
-2000000 10_Networkload (bits/sec)
Simulation time (min)
Figure 35: Analysis of AODV in all experiments
DSR in all experiments

DSR routing protocol was also simulated in all experiments and Figure 36 shows
that DSR has higher throughput results with 40 nodes. The behaviour of DSR
66
(when using UDP) for network load and delay shows consistent results for all
experiments (4, 7, 10, 20, and 40 nodes).
DSR results for all experiments

18000000
16000000 4_Delay (sec)

7_Delay (sec)
14000000
10_Delay (sec)
12000000
20_Delay (sec)
10000000 40_Delay (sec)
6000000
4000000
2000000
0 4_Networkload (bits/sec)
0 2 4 6 8 10 12
Figure 36: Analysis of DSR in all experiments
OLSR in all experiments

OLSR when using UDP was simulated and Figure 37 shows that throughput has
the best results with 40 nodes. Throughput results for 20 nodes is also high and
the behaviour for network load and delay also changes for the other experiments
(refer to figure).
67
OLSR results for all experiments
5000000
4_Delay (sec)
4500000
7_Delay (sec)
4000000
10_Delay (sec)
3500000 20_Delay (sec)
3000000 40_Delay (sec)

1500000
1000000
500000
0
0 2 4 6 8 10 12
Simulation time (sec)
Figure 37: Analysis of OLSR in all experiments
Comparative analysis of all experiments

A comparative analysis of the routing protocols for each experiment is shown in
Figures 38-42 with all testing parameters: throughput; delay; and network load.
All figures show results comparisons of the routing protocols AODV, OLSR,
and DSR for each testing parameter (throughput, delay, and network load) with
4, 7, 10, 20, and 40 nodes. Each color represents the behaviour of a routing
protocol’s average value for each testing parameter performance.
68
Results for experiment with 4 nodes
300000
AODV_Delay
250000
DSR_Delay
200000
OLSR_Delay
150000 AODV_Throughput
100000 DSR_Throughput
OLSR_Throughput
50000
AODV_Networkload
0 DSR_Networkload
0 2 4 6 8 10 12
-50000 OLSR_Networkload
Figure 38: Comparison of AODV, DSR, and OLSR in 4 nodes

600000
500000 AODV_Delay
DSR_Delay
400000
OLSR_Delay
300000 AODV_Throughput
DSR_Throughput
200000
OLSR_Throughput
100000 AODV_Networkload
0 DSR_Networkload
0 2 4 6 8 10 12 OLSR_Networkload
-100000
69

3500000
3000000 AODV_Delay
DSR_Delay
2500000
OLSR_Delay
2000000
AODV_Throughput
1500000 DSR_Throughput
OLSR_Throughput
1000000
AODV_Networkload
500000
DSR_Networkload
0 OLSR_Networkload
0 2 4 6 8 10 12
-500000
70
18000000
16000000
AODV_Delay
14000000
DSR_Delay
12000000
OLSR_Delay
10000000
AODV_Throughput
8000000
DSR_Throughput
6000000
OLSR_Throughput
4000000
AODV_Networkload
2000000
DSR_Networkload
0
OLSR_Networkload
-2000000 0 2 4 6 8 10 12
Nodes Parameters AODV DSR OLSR
4 Throughput (bit/sec) 227,456 231,509.33 231,392
Delay (sec) 0.0004436 0.0002972 0.0003597
Network load (bit/sec) 225,056 230,677.33 225,056
Delay (sec) 0.0007877 0.0003839 0.0007509
Delay (sec) 0.0015 0.000975 0.0008170
71
20 Throughput (bit/sec) 2,881,698.66 876,216 1,206,578.66
Delay (sec) 2.369853 5.765952 2.087846
Network load (bit/sec) 849,645.66 892,045.33 763,914.66
40 Throughput (bit/sec) 12,661,229.33 1,072,970.66 4,153,493.33
Delay (sec) 4.78048 12.351543 3.211828
Network load (bit/sec) 1,156,144 819,378.66 943,986.66
Table 9: Average values of throughput, delay, and network load for AODV, OLSR, and DSR for all experiments
Table 9 gives the average result values of the performance of each routing
protocol after simulation runtime for all five experiments under the metrics:
throughput; delay; and network load. Throughput and network load are
measured in bits per seconds, and delay is measured in seconds.
Conclusion
In this chapter the behaviour of the different protocols were analyzed. In the
following chapter the results will be used to predict the best routing protocol.
72
Chapter 8
FINAL WRITEUP
Introduction
The previous chapter discussed the results analysis of the routing protocols when
using UDP. In this chapter the conclusion based on the project findings is given.
Project conclusion
In this project the performance of the routing protocols AODV, DSR, and
OLSR in terms of the testing parameters: throughput; delay; and network load,
was evaluated for a UDP application – voice calling. Another aspect that was
considered (in the evaluation) was scalability (i.e. increase in number of nodes)
because the simulation results focused on the routing protocols’ performance
when the number of nodes varied.
The motivation of this project was to evaluate the mentioned routing protocols’
performance under the testing parameters in order to find a suitable protocol for
communication. This protocol will be used in disaster events for efficient and
reliable communication when using any type of voice communication device.
The simulation results are provided in graphical and table format. The graphical
format results (Figures 20-34) are retrieved straight from the simulation tool
OPNET whereby the results are generated after a simulation has run. The table
format results (Table 9) represent the average values of each protocol under each
testing parameter so that a comparative analysis can be done from the values.
It was shown that different routing protocols perform differently with different
number of nodes. In the first experiment (4 nodes), DSR performed better in
terms of throughput and delay whilst AODV performed better for network load.
73
As soon as the nodes are increased (7 nodes), OLSR performs better in terms of
throughput, DSR shows little delay, and AODV presents with a low network
load.
In the third experiment (with 10 nodes) OLSR out-performed both DSR and
AODV in terms of throughput and delay. However in terms of network load
AODV outperforms the other protocols. The fourth experiment (20 nodes)
showed that AODV performed better in terms of throughput and OLSR
performed better in terms of delay and network load compared to both DSR and
AODV.
Lastly, the experiment with 40 nodes demonstrated that AODV performed the
best in terms of throughput, OLSR in terms of delay, and DSR in terms of
network load.
To conclude the project results shows that the two routing protocols, OLSR and
AODV, outperformed DSR in peer to peer mesh networks (when using UDP).
For smaller networks DSR gave better results (but the performance degraded
when the network became larger.
But this may not necessarily mean that OLSR and AODV will always
outperform DSR. Different routing protocols have different attributes
depending on the environment (the type of network and traffic type) thus these
need to be considered when deciding on which protocol to apply.
Future work
In future research, different traffic applications which are used in disaster events
(e.g. video communication) should be considered. It will require using simulation
tools that support this type of traffic application e.g. NS2. Furthermore it would
be interesting to get data for a real-life event (that is: apply the testing on a case
74
study) to test if the findings of the simulation would correspond with real-life
results.
Conclusion
This chapter discussed the final conclusion of the project. It has been found that
OLSR and AODV give better performance results depending on the number of
nodes in a simulated network, thus these routing protocols are reliable for peer
to peer networks for disaster events.
75
APPENDICES
APPENDIX A
Probes
Have you ever been involved in a disaster management scenario (e.g. flood,
fire, war, earthquake, etc.)?
IF YES
i. Are you aware of what a Mesh Network is?
ii. If yes have you ever constructed/used one?
iii. What emergency telecommunication networks have you used? Why?
iv. What devices were used during the construction of such a
telecommunication network (e.g. satellite radio, mobile phone,
laptop, etc.)? Why?
v. Did you find the network reliable and efficient? Why?
vi. If not, what other measures did you use to make communication
reliable and efficient?
vii. What problems did you encounter with the telecommunication
network?
viii. What problems did you encounter with the devices used?
76
IF NO
i. Are you aware of what a Mesh Network is?
ii. If yes have you ever constructed/used one?
iii. What telecommunication network do you propose you would use if a
disaster were to occur in your area
iv. What communication devices would you use
v. How would you go about constructing an efficient
telecommunication network?
vi. How would you be able to assess if the constructed network will be
reliable and efficient?
77
APPENDIX B
Data from questionnaires
Mr. James Chilonga (Mobile & VSATs Network Administrator at

Broadband Network, Malawi)
No, we have not experienced one – but we have a disaster routine plan.
Are you aware of what a Mesh Network is?
Yes I am
If yes have you ever constructed/used one?
Yes – I was a junior network analyst for a certain project

What telecommunication network do you propose you would use if a
disaster were to occur in your area?
A wireless mesh network can be seen as a special type of wireless ad-hoc network. Wireless
mesh network often have more planned configuration, and may be deployed to provide dynamic
and cost effective connectivity over a certain geographic area.
What communication devices would you use?
Radio Antenna, servers

How would you go about constructing an efficient telecommunication
network?
Wireless mesh architectures infrastructure is, in effect, a router network minus the cabling
between nodes. It's built of peer radio devices that don't have to be cabled to a wired port like
traditional WLAN access points (AP) do. Mesh architecture sustains signal strength by
breaking long distances into a series of shorter hops. Intermediate nodes not only boost the
signal, but cooperatively make forwarding decisions based on their knowledge of the network, i.e.
perform routing. Such architecture may with careful design provide high bandwidth, spectral
efficiency, and economic advantage over the coverage area.
78
How would you be able to assess if the constructed network will be reliable
and efficient?
This type of infrastructure can be decentralized or centralized; both are relatively inexpensive,
and very reliable and resilient, as each node needs only transmit as far as the next node. Nodes
act as routers to transmit data from nearby nodes to peers that are too far away to reach in a
single hop, resulting in a network that can span larger distances. The topology of a mesh
network is also reliable, as each node is connected to several other nodes. If one node drops out of
the network, due to hardware failure or any other reason, its neighbors can quickly find another
route using a routing protocol
Mr. Clint Prince (Assistant Operations Administrator at Campus

Protection Services, University of the Western Cape)
No.
Are you aware of what a Mesh Network is?
No
If yes have you ever constructed/used one?
N/A
What telecommunication network do you propose you would use if a
disaster were to occur in your area?
The Information Communication Services (ICS) department has a system called GV Chase
System that sends out text messages (SMSs) to staff, students and workers on campus when a
disaster incident occurs to warn them of the dangers that might affect them.
What communication devices would you use?
Mobile phones and servers

How would you go about constructing an efficient telecommunication
network?
Construction of the GV Chase System is still under the beta phase. Since a disaster incident
has never occurred on campus, the ICS department has not had the chance to use it.
79
How would you be able to assess if the constructed network will be reliable and
efficient?
I would recommend that a disaster response drill could occur on campus to test the effectiveness of
the system in preparation for any future disaster.
80
APPENDIX C
Term 1 Project plan

Meeting dates Tues 1st Feb Tues 8th Feb Thurs17th Feb Mon 21st Feb Mon 28th Feb Mon 7th March Mon 14h March
& times
/Tasks 10h00 12h00 10h00 10h00 10h00 10h00 10h00
Identify users to Identify users to Identify users to Identify someone Mock Presentation Presentation –
interview interview or give a interview or give a (Writing Centre) to on Tuesday Tuesday another
Comments questionnaire to. questionnaire to. read your doc. & mock presentation &
organise someone the real one on
Note questionnaire or Note questionnaire for checking the Wednesday! (09h00
probes and add as or probes and add presentation. – 11h00
appendix. as appendix.
Thesis See that you

Create document
Document See previous week! understand how to o Check Index and Finalise write-up.
using Honours
Complete the thesis use the Styles and add indexes
Project Guidelines and start with Let someone
outline (using how to compile an
from the website as RAD. o Bibliography – at proof read your
Honours Project Index, the Table of
well as Thesis doc least 5 entries. document!
Guidelines) Contents, List of
from Word
Figures etc.
URD Fill in headings

Fill in headings Finnish with URD
Look at the write-up. Interview Complete URD
questionnaire. Start write-up of stakeholders.
URD.
RAD n/a Fill in headings Start write-up of Write-up RAD Complete RAD
81
RAD.
Literature Familiarising yourself

Survey with your topic, and Add all literature
how it's implemented. Read and explore. found to your
some more – use bibliography – use Keep on reading
Read and explore. Google Scholar.. the Harvard
literature on your Notation
topic.
Presentation/ Write one paragraph Write one paragraph

deliverable that describes what that describes what
Use thesis to
you want to do, and you want to do, and
Prepare slides for Mock presentation
why you want to do it why you want to do it
mock presentation
This will be used as This will be used as
the abstract the abstract
Website Put plan onto

Ask Frieslaar website
Ask Frieslaar
Ask Frieslaar
intelligent.networking Ask Frieslaar about the server. Add URD & RAD
intelligent.networking @gmail.com See to website
intelligent.networkin Add URD to
@gmail.com
Create website g@gmail.com See website
Create website
82
Term 2 Project plan
Tasks 28th t March 4th April 11th April 18th April 25th April 2th May 9th May 16th May 23rd May
Combined Combined Monday a Monday holiday Combined
Comments
meeting meeting holiday Combined meeting meeting
Create 3 new Complete Start write up of Complete Complete Complete write-up of Complete write Complete write
chapters with editing. Start the OOA & write-up of write up of OOA. up of OOD up.
Thesis Document
subheadings with the write complete the OOA. OOD Re-organise the 20th May: Hand
for UIS, OOA up of the UIS UIS write-up document to reflect in completed
and OOD – an experiment rather document to
see p3 than a program. supervisor
OOA Read through Start with Complete Analysing the RAD to

the analysing the OOA create OOA
documentatio RAD to create
n of this OOA
OOD carefully and Start write up Complete Start write up of Complete OOD
see that you of OOD OOD OOD
understand it! Start the writing up of
the planning of the
data collection and
experiments that will
be done.
83
UIS Start with User Complete UIS Update changes to If needed -
Interface UIS update changes
Specification to UIS
GUI & Start with the Program GUI/ Program Program GUI/ Complete GUI/ Finalise GUI/
prototype planning of the prototype GUI/ prototype prototype prototype
prototype prototype Plan and start with a
pilot study
Look at Look at
previous previous
Other
projects projects
Prepare slides Mock

for mock Presentation
Presentation
presentation on the 24th &

(24th May) – to Presentation
view on the on the
23rd May 25thMay
Web- Update web Check and update Put new plan,
site site thesis
&presentation
on web site
84
Term 3 Project plan
Maggie Bila
th th th nd th th th th th th th
Tasks 12 July 19 July 26 July 2 Aug 9 Aug 16 Aug 23 Aug 30 Aug– 5 Sept 5 Sept 12 Sept
Thesis Finalise the Finalise the Make changes to object’s pseudo code as you develop the software, Finalise
Document editing of the editing of the document all changes etc. in the code&start on the User’s guide (User’s Documentation and
th
documentation documentation - Guide a deliverable for the next term only!) hand in on the 27
& editing
Update any
changes to the
design – e.g.
objects
Planning of Reconsider how to represent the data – Write the skeleton of your Chapter 5 – how you do
chapters planning of Chapters. the setting up and what scenarios you will be
considering. And write the results of some
experiments that you have done in Chapter 6 – no
discussions yet.
Re-visit the GUI Check the GUI Re-design parts of the GUI or the whole GUI or just change Replace Finalise GUI
and make and see if you it for the time being to red – so that you can change it later screenshots with
changes or are happy that screenshots of
redesign it deals with all the current
the options program (it will
have changed)
Maggie revisit
the brief.
Scenarios Plan scenarios and determine what aspects will be considered in the scenario and tested
(simulation)
Create & Create & populate (add a few
populate data references to) the MySQL
database database or put together files
to be used in programme
85
th
Programming Plan the Program Program Programming Program 4 Finalise
st st nd
task approach by 1 1 &2 task/module programming &
breaking task task/modul task/mod /object testing
into objects or e /object ule
modules to /object
program
Testing and Read about Read about Read Decide on a
refining with a MySQL software & about subset of
basic data set database tools you wish software testing data
and decide to implement & tools
on its you wish Testing and refining
structure or to
if you use implemen
files how it t
will be used.
th h
Mock 6 Present 14
Prepare for
Presentation Sept. September
presentation
Present
Website Update NB Update NB Update NB Update NB
86
Term 4 Project plan
th st th th th th st
Tasks for the 20 Sept 27 Sept 4 October 11 October 18 Oct 25 Oct 1 Nov
week of the nd
Presentation on the 2
rd
Thesis Finalise the Complete all editing of 3 Start by identifying all the tasks that the program Finalise user’s Hand in
th
Document editing of the term’s must be able to do - write the User’s guide guide & Thesis documentation on Monday 8
documentation documentation. documentation November
. Ask
colleague or
someone to
edit /proofread
it!!
Programming Revise programme & add what is still outstanding Revise Revise programme
tasks not programme &
completed And make changes to thesis doc to reflect this! create Finalise Installation disc.
Mr Bila please re-think your project and highlight what you have and plan to achieve installation
disc.
Design Test Read up about Add the theoretical part of Design test suites
suites evaluation of evaluation to your
programs – get documentation. Edit the Choose your “users”
ideas of how you chapters where you Questionnaires etc.
want to do referred to it initially.
evaluation. Add Ethical aspects
to
documentation.
Edit the chapters
where you
referred to it
initially.
87
Execute tests, Execute tests & keep record of it
revise code or by writing it up in Thesis doc.
even project Create graphs that can be
design included in Thesis doc.
st
Presentation Mock Presentation on the 1
Nov.
Prepare for presentation and
install on relevant computers.
Test if it works!!
Website Update NB Update NB Final Update

NB
88
BIBLIOGRAPHY
1. Ali, S., & Ali, A. (2009). Performance Analysis of AODV, OLSR, and DSR
in MANET. Retrieved 01 18, 2011, from essay.se: http://www.essay.se
2. Arnold, J. L., Levine, B. N., Manmatha, R., Lee, F., Shenoy, P., Tsai,
M. C., et al. (2004, 07-09). Information-sharing in out-of-hospital disaster
response: The future role of information technology. Retrieved 08 15, 2011,
from Prehospital and Disaster Medicine:
http://pdm.medicine.wisc.edu
3. Bille.R. (2010). Action without change? On the use and usefulness of pilot
experiments in enviroment mangement. Retrieved 05 04, 2011, from
S.A.P.I.E.N.S.: www.wikipedia.com
4. Boehm, B. (1986, 08 14). A Spiral Model for software development and

enhancment. Retrieved 07 18, 2011, from ACM, SIGSOFT:
www.wikipedia.org
5. DiMarco, C. (2011, 01 11). Utilising Voice Broadcasts in Disaster Scenarios.

Retrieved 03 12, 2011, from TMCnet: www.tmcnet.com
6. Haralambos, H. (2000). Sociology: Themes and Perspectives. Retrieved 05 04,

2011, from Wikipedia: www.wikipedia.com
7. IEEE. (1999). Wireless LAN Medium Access Control (MAC) and Physical
Layer Specifications. Retrieved 06 2011, from IEEE Std 802.11:
www.ieee.com
89
8. Kim, J., Vinay, S., & Stephen, B. (2008, 04 19). Realistic Simulation of
Urban Mesh Networks. Retrieved 03 15, 2011, from
http://www.eecis.udel.edu
9. Kumar, V. (2009, 07). Simulation and comparison of AODV and DSR in

routing protocols in MANETs. Retrieved 05 15, 2011
10. Kurose, J. F., & Rose, K. W. (2010). Computer Networking: User Datagram
Protocol. Retrieved 05 10, 2011, from Wikipedia:
en.wikipedia.org/wiki/User_Datagram_Protocol
11. MacVittie, L. (2011, 06 11). IPv4 to IPv6 switch: When protocols collide.
Retrieved 07 12, 2011, from em.wikipedia.org
12. Oliverz-Giles, N. (2011, 03 11). Google deploys Person Finder after Japan
earthquake, tsunami leaves hundreds dead. Retrieved 03 12, 2011, from
http://latimesblogs.latimes.com
13. OPNET Technologies Inc. (2009). OPNET Support Center. Retrieved

06 15, 2011, from www.opnet.com
14. Preparedness, Y. C. (2008, 12). Tactical Communication. Retrieved 02 03,

2011, from Public Health: http://publichealth.yale.edu/ycphp
15. Rahman, A., Islam, S., & Talevski, A. (2010). Performance measurement of
various routing protocols in ad-hoc network. Retrieved 02 21, 2011
16. Ravikumer, Y., & Chittamuru, S. K. (2010, 06). A Case Study on

MANET Routing Protocols over HTTP and TCP. Retrieved 01 25, 2011,
from essay.se: http://www.essay.se
90
17. Townsend, M. A., & Moss, L. M. (2005, April). Telecommunications
Infrastructure in Disasters: Preparing cities for crisis communication. Retrieved
02 08, 2011, from New York University:
http://hurricane.wagner.nyu.edu
18. Usop, N. M., Abdullah, A., & Abidin, A. F. (2009). Performance

Evaluation of AODV, DSDV & DSR Routing Protocol in Grid
Environment. IJCSNS International Journal of Computer Science and
Network Security, VOL.9 No.7, 7.
91
INDEX
A
ACK, viii, 9
N
Acknowledgment message. See ACK Network Interface Card. See NIC
Ad hoc on demand distance vector. See NIC, viii, 2
AODV NS2, viii, 75, See Network simulator
Ad hoc On Demand Vector. See AODV
ADOV, 19, 66
AODV, viii, 3, 4, 8, 9, 11, 16, 23, 25, 28, 29, O
31, 35, 40, 41, 51, 56, 61, 66, 67, 69, 74, OLSR, viii, 3, 4, 8, 10, 11, 16, 19, 23, 25, 28,
75, 76 31, 35, 40, 41, 51, 56, 61, 66, 68, 69, 74,
75, 76
D OPNET, ix, 7, 12, 14, 15, 16, 18, 19, 22, 23,
24, 31, 32, 42, 46, 51, 74
DSR, viii, 3, 4, 8, 9, 11, 16, 19, 23, 25, 28, 29, Optimized link state routing. See OLSR
31, 35, 40, 41, 49, 51, 56, 61, 66, 67, 69, Optimized Link State Routing. See OLSR
74, 75 Optimized Network Evaluation Tool. See
Dynamic source routing. See DSR OPNET
Dynamic Source Routing. See DSR Optimized network tool. See OPNET
OSI, 10
F
File Transfer Protcol. See FTP
P
FTP, viii PDA, ix, 1
Personal Digital Assistant. See PDA
H
HTTP, viii
R
Hypertext Transfer Protocol. See HTTP RERR, ix, 9
Route error message. See RERR
I Route reply message. See RREP
Route request message. See RREQ
IP, 10 RREP, ix, 9
RREQ, ix, 3, 9
M
S
MAC, viii
MANET, viii, 11 Short Message Service. See SMS
Media Access Control Adress. See MAC SMS, ix, 5
MID, viii, 10
Mobile Ad hoc Network. See MANET
MP3, viii, 1, 5
T
MPEG-2 Audio Layer 3. See MP3 TC, ix, 10
Multiple interface declaration. See MID TCP, ix, 10
92
Topology control. See TC
Transmission Control Protocol. See TCP
V
Voice over IP. See VoIP
U VoIP, ix
UDP, ix, 3, 7, 8, 10, 11, 12, 16, 19, 23, 25, 35,
40, 41, 46, 56, 61, 67, 68, 74, 75
W
User datagram protocol. See UDP WiMAX, ix
User Datagram Protocol. See UDP Worldwide Interoperability for Microwave
Access. See WiMAX
93

Mini Thesis PDF

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

Mini Thesis PDF

Uploaded by

Copyright:

Available Formats

EVALUATING THE

A thesis submitted in partial fulfillment of

BSc. Honours Computer Science

University of the Western Cape

Date: November 1, 2011

Evaluating the performance of Mesh network protocols in disaster scenarios

Supervisor: Professor I.M. Venter

Department of Computer Science

Table of Contents .......................................................................................... i

TABLE 1: OVERVIEW COMPARISON OF MESH ROUTING PROTOCOLS 9

Acknowledgment message (ACK). A message used to send and

Ad hoc on demand vector (AODV). A routing protocol that generates a route

Dynamic source routing (DSR).A self-organizing and self-configuring protocol

Hypertext transfer protocol (HTTP). A protocol that communicates between

Media access control (MAC). A unique identifier assigned to network

Mobile ad hoc network (MANET).A self-configuring network of mobile

Multiple interface declaration (MID).A message that is sent to communicate

Moving pictures experts group (MPEG-2 audio layer 3 / MP3). A common

Network interface card (NIC). A computer component that connects a

Network simulator version 2 (NS2). A discrete event simulator targeted at

Personal digital assistant (PDA).A mobile device that functions as a personal

Route request message (RREQ). A communication message that is sent in

Request error message (RERR). A message that is sent if the destination

Short message service (SMS). The text communication service component of

Transmission control protocol (TCP). A protocol in the transport layer

User datagram protocol (UDP). A protocol often used in videoconferencing

Voice over internet protocol (VoIP).A transmission technology for delivery of

Worldwide interoperability for microwave access (WiMAX).A

USER’S REQUIREMENTS DOCUMENT

Sketching the background

Performing an evaluation analysis on Mesh network protocols, could help with

Performance of mesh networks

Advantages of mesh networks

Disadvantages of mesh networks

User’s view of the problem

In disaster recovery scenarios reliable communication is of utmost importance.

Two communication difficulties have been identified voice and video

Brief description of users problems

What is expected from the testing of the performance of the protocols in

OVERVIEW OF THE PROTOCOLS

Mesh routing protocols and transport protocol

Routing Type of Frequency of Worst case Multiple

Routing DSR Reactive Source As needed Full flooding Yes

AODV Reactive Hop-by-hop As needed Full flooding No

OLSR Proactive Hop-by-hop Periodically Pure link state No

Transport UDP Connectionless Host-to-host Periodically Congestion No

Table 1: Overview comparison of mesh routing protocols

Ad hoc on-demand distance vector routing protocol (AODV)

Dynamic source routing protocol (DSR)

Optimized links state routing protocol (OLSR)

User datagram protocol (UDP)

UDP is often used in videoconferencing applications or computer games

Why these protocols?

Why use the OPNET simulation tool?

View and Analyze Collect Statistics

Figure 3: OPNET Workflow Model.

SIMULATION ENVIROMENT ANALYSIS

Brief description of complete simulation environment analysis

Figure 4: OPNET Workflow

Brief description of a scenario

They need to set up a wireless ad hoc network with the 4 communication

No. of Routing Other Performance Simulation Mobility rate Simulation

1 4 AODV UDP Throughput, 2000m x 2000m 10meters/sec 10min