Skip to main content

MANET Routing Protocols using ns3


Download the Code here:

Check the Video for Detailed instruction and how to use the plots, graphs and source code.

Comparison of MANET routing Protocols

3. DSR
using NS3 (Network Simulator 3)
B.Tech, M.Tech, PhD...
1. Reactive Vs Proactive routing
2. PErformance comparison of MANET protocols
3. AODV Vs DSDV Comparison and also at my channel.

What Version: ns-3.29
My Ubunut OS is: Ubuntu 18.04

This file we are going to use for our simulation:

Once you under stand the code, now lets run this example

Step 1: Copy the above file in to ~ns-3.29/scratch/ folder

Step 2: Understand this code.
Step 3: Run this code
Open the terminal, Go to ns-3.29 and run the following command

$] ./waf --run scratch/manet-routing-compare

enable the following header file

#include "ns3/flow-monitor-helper.h"

four files got generated

Number of nodes: 50
No of sinks : 10
Mobility Model:
Propagation Model: ConstantSpeedPropagationDelay
Propagation Loss Model: Friis
Position Allocator: RandomRectangularPositionAllocator
Mac: AdhocWifiMAC
Mac Standard: 802.11B
Bps: 2Kbps
Total Simulation Time: 200 seconds
Node speed: 20m/s
Node pause time: 0
Protocol: AODV

To run the flowmon file, use of is advised.

to run the file, use the following

$] python AODV.flowmon

Find the python script here.

#beginning of Script

from xml.etree import ElementTree as ET
import sys
import matplotlib.pyplot as pylab
for flow in et.findall("FlowStats/Flow"):
for tpl in et.findall("Ipv4FlowClassifier/Flow"):
if tpl.get('flowId')==flow.get('flowId'):
if tpl.get('destinationPort')=='654':

if rxPackets==0:

pylab.xlabel("Flow Bit Rates (b/s)")
pylab.ylabel("Number of Flows")

pylab.xlabel("No of Lost Packets")
pylab.ylabel("Number of Flows")

pylab.xlabel("Delay in Seconds")
pylab.ylabel("Number of Flows")


#End of Script

We can plot using gnuplot (through CSV File) and matplotlib(Flow monitor)
All these examples, runs for 50 nodes

whenever you compare protocols,have three different sets of node numbers

10 nodes, 50 nodes, 100 nodes and compare them with various metrics as mentioned above...

To analsyse .tr files, we need to use tracemetrics
its a jar file that can be run using

$] java -jar tracemetrics.jar

Hope you would have got an idea on how to simulate wireless protocols using ns3.


Popular posts from this blog

Installation of ns2 in Ubuntu 18.04 - NS2 Tutorial # 2

ns2 installation in Ubuntu 18.04

Unlike earlier version of ubuntu, the 18,04 version does not support the direct installation of ns2, because of the latest C/C++ compiler (7.x), but the latest C/C++ compiler supported by ns2 is gcc-4.8. So this post and video tells you about the installation of gcc-4.8 version, then ns2 and then configuration.
Here are the steps, Please follow the video given below for more information.

$] tar zxvf ns-allinone-2.35.tar.gz

$] sudo apt install gcc-4.8 g++-4.8

if you have installed ubuntu just now, you can try these commands also in the beginning

$] sudo apt update
$] sudo apt install build-essential autoconf automake libxmu-dev

$] cd ns-allinone-2.35/ns-2.35
$] gedit
$] gedit linkstate/ls.h

line number 137, change erase to this->erase
$] cd ..
$] ./install

export PATH=$PATH:/home/pradeepkumar/ns-allinone-2.35/bin:/home/pradeepkumar/ns-allinone-2.35/tcl8.5.10/unix:/home/pradeepkumar/ns-allinone-2.35/tk8.5.10/unix

ns3 installation in Ubuntu 16.04

This post serves the installation instructions of ns3 in ubuntu 16.04 version. Some of my students are working in ns3, this post will benefit them in installing ns3.

OS Used: Ubuntu 16.04.4
ns3 version: ns3 version 3.27

The same procedure will be applied for OS like Debian, Linux Mint.

Fresh installation of Ubuntu 16.04 Let you try the fresh installation of ubuntu in your hard disk along with windows.
Installation of ns3 dependencies ns3 needs so many dependencies, developmental libraries, drivers, etc. so install all those
$] sudo apt update  $] sudo apt upgrade
$] sudo apt-get install build-essential autoconf automake libxmu-dev python-pygoocanvas python-pygraphviz cvs mercurial bzr git cmake p7zip-full python-matplotlib python-tk python-dev python-kiwi python-gnome2 python-gnome2-desktop-dev python-rsvg qt4-dev-tools qt4-qmake qt4-qmake qt4-default gnuplot-x11 wireshark
The above command make take some time to download, compile and install it, Be Patient. 
Installing ns3 Download the…

AWK Scripts for NS2 to process data from Trace Files

AWK Scripts are very good in processing the data from the log (trace files) which we get from NS2. If you want to process the trace file manually, here is the detailHere is a sample of trace file from NS2 (However ns2 supports a new type of trace file also), but this post will make you understand the old trace format only.r 0.030085562 _0_ MAC  --- 0 message 32 [0 ffffffff 1 800] ------- [1:255 -1:255 32 0] r 0.030110562 _0_ RTR  --- 0 message 32 [0 ffffffff 1 800] ------- [1:255 -1:255 32 0] s 1.119926192 _0_ RTR  --- 1 message 32 [0 0 0 0] ------- [0:255 -1:255 32 0]
AWK Scripts are very good in processing the data column wise. For example
the first column in the above trace file represents r, s which indicates receive, sent respectively. If we want to trace the entire r and s alone from this trace file we can represent it as $1
$1 represents ACTION
$2 Time
$3 Node ID
$4 Layer
$5 Flags
$6 seqno
$7 type
$8 Size
$14 Energy (if the network nodes includes EnergyModel)To run the awk script in Li…