Class 7

Data Collection and Analysis in
Experimental Research
With the Applications of NS3 Simulator
Objectives for this session
 To Introduce simulation and its Advantages

 To discuss overview of few popular Simulation tools
 To demonstrate the open source tool NS3 .
 To show the architecture of ns3 source code and building it
 To view the implementation techniques by running few programs
 To Demonstrate a sample research related problem & solutions.
 To Analyze Data and Report the results.
 To Conclude and Summarize.
Presentation Sequence and Topics
Introduction
Overview of Simulation Tools
NS3
Practical - Experiments
Structure
Code Building
Executing Programs
Research
Data Collection using programs
Data Analysis
Conclusion and Summary

Introduction
Simulation is the imitation of the operation
of a real-world process or system.
Using simulation one can design, simulate, verify,
and analyze the performance of different
networks protocols.
When to use?
The uses of simulation in business are varied.
If we cannot conduct experiments on a real system
then we can use simulations.
Who uses?
Academic researchers,
Industrial developers, and
Quality Assurance (QA) Units.
With their help, one can design different network

topologies using various types of nodes such as
end-hosts, hubs, network bridges, routers, optical link-
layer devices, and mobile units.
Why we use Simulation ?
Computer models are cheaper to setup than

alternative methods
We can predict what will happen in a system, eg

building a prototype.
We can make alterations and quickly see the

outcomes.
This allows the designer to determine the correctness and

efficiency of a design before the system is actually
constructed.
Overview of Simulation
Tools
Commercial - OPNET, QualNet
Open source - NS2, NS3, OMNeT++
OPNET, QUALNETa full graphical user interface
Prepared by Prathap.M
NS3
NS3 is designed to replace the current popular NS2 .
However, NS3 is not an updated version of NS2 since

that NS3 is a new simulator and it is not backward-
compatible with NS2.
Developed from scratch with script support of C++

and Python.
NS 3' s features
Practical - Experiment
Introduction
Organization of the Stack

Research
Formulate Model/ Design - Protocols for
comparision
DSDV, DSR, AODV
Programs related to these protocols
Table showing their filename, xml and csv files.

Data Collection
Can be collected as .csv files or .xls files
Collect common data each file

and merge files
Data Analysis -
1. Factor: Pkt delivery ratio
Total Pkt delivery Average Pkt delivery
20 nodes 20 nodes
AODV 2281 11.405

DSR 2160 10.8
DSDV 1338 6.69
Cont…
.
A final step in the data analysis is to present the
results in tables or figures and interpret the results
from the statistical test.
18
Scenario 1
Average Pkt delivery Average Pkt delivery Average Pkt delivery
20 Nodes 100 Nodes 250 Nodes
AODV 11.405 x y
DSR 10.8 x-2 y-4
DSDV 6.69 x-3 y-5

Scenario 2
End to End End to End End to End
Delay Delay Delay
20 Nodes 100 Nodes 250 Nodes
AODV 3.40 x y
DSR 4.1 X +2 Y+ 4
DSDV 4.2 X+3 Y+ 5

Scenario 3 Packet delivery ratio of Mobility
Model- Ref: value is represented in each bar
of the chart
RPGM - Reference Point Group Mobility

RW - Random Waypoint
GM - Gauss Markov
MG - Manhattan Grid
Packet delivery Ratio
20 nodes - In this scenario RPGM Outperforms
When the node is increased to 100 RW outperforms
20 nodes - RPGM Outperforms

100 nodes – RW Outperforms
Therefore RPGM is good for low sized networks

Whereas RW for medium networks
Factor 2 : End to End delay
Simulation result of end to end delay of protocols
DSR extends too high

DSDV Medium
AODV is less
Since it is delay , the delay is minimum in AODV and hence

AODV outperforms both in packet delivery and less delay.
Conclusion and
Summary
Summary
We discussed a brief introduction of network
simulation.
We viewed experiment, results and analyzed it.
If necessary, we can export the excel files and

by using SPSS and analyse.
The open sourced and well designed embedded Object-
oriented programming feature will be appealing to the
academia and industry in the long run.
It seems that the that simulator will achieve popularity based

on easier for the users to learn and to use, the more
organizations participate in, the more specialized in the
documentation, the more development and popularity.
Summary
 Step 1. Identify the problem.
 Step 2. Formulate the problem.
 Step 3. Collect and process real system data. (Using Real
time packet and traces.)
 Step 4. Formulate and develop a model.
 Step 5. Validate the model.
 Step 6. Document model for future use.
 Step 7. Select appropriate experimental design.
 Step 8. Establish experimental conditions for runs.
 Step 9. Perform simulation runs.
 Step 10. Interpret and present results.
 Step 11. Recommend further course of action.
References
Jianli Pan, Research Paper titled “A Survey of Network
Simulation Tools: Current Status and Future Developments”
[WANsim] WAN simulators and emulators, â€œ
http://www.wan-sim.net/
[OPNET] OPNET Modeler, http://www.opnet.com/
[NS2] NS2 official website, http://www.isi.edu/nsnam/ns/
[NS2-wiki] NS2 resource webpage,
http://nsnam.isi.edu/nsnam/index.php/Main_Page
[NS3] NS3 official website,
http://www.nsnam.org/documents.html
[OMNeT] OMNeT++ official website,
http://www.omnetpp.org/
[REAL] REAL 5.0 simulator overview,
http://www.cs.cornell.edu/skeshav/real/overview.html
.
Arsalan Iqbal and Sameer Liaqat Ali Khan Performance Evaluation of Real Time
Applications for RIP, OSPF and EIGRP for flapping links using OPNET Modeler
International Journal of Computer Networks and Communications Security VOL. 3,
NO. 1, JANUARY 2015, 16–26 www.ijcncs.org E-ISSN 2308-9830 (Online) / ISSN
2410-0595 (Print)
Network, Performance Evaluation International Journal Elsevier (57) , 141-161, 2004.
www.nsnam.org
THANK YOU

Class 7

Uploaded by

Copyright:

Available Formats

Class 7

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

Class 7

Uploaded by

Copyright:

Available Formats

Data Collection and Analysis in

 To Introduce simulation and its Advantages

Conclusion and Summary

With their help, one can design different network

Computer models are cheaper to setup than

We can predict what will happen in a system, eg

We can make alterations and quickly see the

This allows the designer to determine the correctness and

OPNET, QUALNETa full graphical user interface

However, NS3 is not an updated version of NS2 since

Developed from scratch with script support of C++

Organization of the Stack

DSDV, DSR, AODV

Programs related to these protocols

Table showing their filename, xml and csv files.

Can be collected as .csv files or .xls files

Collect common data each file

1. Factor: Pkt delivery ratio

Total Pkt delivery Average Pkt delivery

AODV 2281 11.405

Average Pkt delivery Average Pkt delivery Average Pkt delivery

20 Nodes 100 Nodes 250 Nodes

DSR 10.8 x-2 y-4

DSDV 6.69 x-3 y-5

20 Nodes 100 Nodes 250 Nodes

DSDV 4.2 X+3 Y+ 5

RPGM - Reference Point Group Mobility

20 nodes - In this scenario RPGM Outperforms

When the node is increased to 100 RW outperforms

20 nodes - RPGM Outperforms

Therefore RPGM is good for low sized networks

DSR extends too high

Since it is delay , the delay is minimum in AODV and hence

We viewed experiment, results and analyzed it.

If necessary, we can export the excel files and

It seems that the that simulator will achieve popularity based

Network, Performance Evaluation International Journal Elsevier (57) , 141-161, 2004.

You might also like