Considerable Detection of Black Hole Attack and Analyzing its Performance on AODV Routing Protocol in MANET (Mobile Ad Hoc Network)

Full text

Considerable Detecon of Black Hole Aack and Analyzing its Performance on
AODV Roung Protocol in MANET (Mobile Ad Hoc Network)
Ashok Koujalagi*
Department of Computer Science, Basaveshwar Science College, Bagalkot, Karnataka, India
*Corresponding author: Ashok Koujalagi, Department of Computer Science, Basaveshwar Science College, Bagalkot, Karnataka, India, E-mail:
koujalagi.ashok@gmail.com
Received date: May 30, 2018; Accepted date: June 14, 2018; Published date: June 22, 2018
Citaon: Koujalagi A (2018) Ulizaon of edge posion for digital image watermarking using discriminant analysis. Am J Compt Sci Inform Technol
Vol.6 No.2:25
Copyright: © 2018 Koujalagi A. This is an open-access arcle distributed under the terms of the Creave Commons Aribuon License, which
permits unrestricted use, distribuon, and reproducon in any medium, provided the original author and source are credited.
Abstract
A Mobile Ad hoc Network is an aggregaon of mobile
terminal that form a volale network with wireless
interfaces. Mobile Ad Hoc Network has no any central
administraon. MANET more vulnerable to aacks than
wired network, as there is no central management and no
clear defence mechanism. Black Hole Aack is one of the
aacks against network integrity in MANET. In this type of
aack all data packets are absorbed by Black Hole node.
There are lots of techniques to eliminate the black hole
aack on AODV protocol in MANET. In this paper a soluon
named Black Hole Detecon System is used for the
detecon of Black Hole aack on AODV protocol in MANET.
The Black Hole Detecon System considered the rst route
reply is the response from malicious node and deleted, then
the second one is chosen using the route reply saving
mechanism as it come from the desnaon node. We use
NS-2.35 for the simulaon and compare the result of AODV
and BDS n soluon under Black Hole aack. The BDS
soluon against Black hole node has high packet delivery
rao as compared to the AODV protocol under Black hole
aack and it’s about 46.7%.The soluon minimize the data
loss and decrease the average Jier 5% and increase the
throughput.
Keywords: MANET; AODV; Blackhole AODV; BdsAODV
Overview to Manet (Mobile Ad Hoc
Network)
Mobile Ad-Hoc Network is an autonomous system where two
or more wireless devices or terminals that has the capability to
communicate with each other without of any centralized
administrator or xed network infrastructure. Mobile nodes can
dynamically form a network to exchange informaon without
the help of any central administraon. MANET are self-organized
networks. In MANET mobile nodes are accountable for
dynamically discovering other nodes to communicate [1]. Here
networks funcons like data forwarding, roung, and network
administraon are carried out synergec by all available nodes.
In Mobile ad hoc network nodes that are in the radio range can
communicate directly, but the nodes that are out of the range
can communicate through the intermediate nodes. Nodes are
free to move randomly while being able to communicate with
each other without the help of an exisng network
infrastructure. Here mobile node operates not only as an anchor
but also as a router for transferring data for other mobile nodes
in the network. MANETs are suitable for the situaons where
any wired or wireless infrastructure is damaged or destroyed [2].
Roung Protocols in Manet
Roung protocol in MANET can be classied into three
categories:
Proacve roung protocol (table-driven roung
protocol)
In this Roung Protocols, in the network each node must keep
up-to-date roung tables. When the network topology changes
every node in the network propagates the update message to
the network to maintain a reliable roung table. The
disadvantages of this roung protocol are that the periodically
updang the network topology increases bandwidth overhead
and many redundant route entries to the specic desnaon
unnecessarily take place in the roung tables. The advantage is
that, if any aacker node joined in network cannot easily aack
the network for geng data. Desnaon Sequenced Distance
Vector (DSDV) and Opmized Link State Roung (OSLR), Wireless
Roung Protocol (WRP), Global State Roung (GSR) are most
familiar types of roung protocols of proacve roung protocol
[3].
Reacve roung protocol (on-demand roung
protocol)
In reacve roung protocol route tables are created when
required and are not maintained periodically. The source node
propagates the route request packet to its neighbors when it
wants to connect to a desnaon node. The neighbors of the
source node receive the broadcasted request packet and
forward the packet to their neighbors unl source node’s
Editorial
iMedPub Journals
http://www.imedpub.com/
DOI: 10.21767/2349-3917.100025 ISSN 2349-3917
Vol.6 No.2:25
2018
© Under License of Creative Commons Attribution 3.0 License | This article is available from: https://www.imedpub.com/computer-science-and-information-
technology/
1
American Journal of Computer Science and Information Technology
ISSN 2349-3917

desnaon is found. Aer receiving the request the desnaon
node sends a route replay packet to the source node through
the shortest path. The path is maintained in the route tables of
the nodes through shortest path unl the route is no longer
needed. This roung protocol is easily aected by the malicious
node. Ad-hoc On-demand Distance Vector (AODV), Dynamic
Source Roung (DSR) is most familiar roung protocols of acve
roung protocol [3].
Hybrid roung protocol
The hybrid roung protocol combine the proacve and
reacve roung protocols. It is discovered using the advantages
of proacve and reacve roung protocols. Hybrid roung
protocol uses the proacve roung protocol in the case of intra-
domain roung and uses the reacve roung protocol in the
case of inter-domain roung. Zone Roung Protocol (ZRP),
Temporally-Ordered Roung Algorithm (TORA) is most familiar
types of roung protocol of hybrid roung protocol [4].
Ad-Hoc On-Demand Distance Vector
(AODV) Roung Protocol
In MANET AODV roung protocol is an on-demand roung
protocol used to nding a route to the desnaon. All mobile
nodes work cooperavely to nding route to the desnaon
using the control messages of roung protocol. In AODV roung
protocol routes are maintained just as long as it is needed.
AODV roung protocol use the desnaon sequence number for
each route entry which is a disnguishable feature from other
roung protocol. In AODV roung protocol the roung table
stores the desnaon address, next-hop address, desnaon
sequence number and lifeme. In this, when a node wishes to
send a packet to some desnaon, it checks its roung table to
determine if it has a pre-established route to the desnaon. If
it has a pre-established route to the desnaon, it forwards the
packet to next node. If it has not a pre-established route to the
desnaon it launches a route discovery process. For
establishing a route to the desnaon the AODV protocol use
the Route Requests (RREQs), Route Replay (RREPs), Route Errors
(RERRs) control messages. The source node broadcasts an RREQ
message when it wants to established a communicaon with the
targeted node. This RREQ message is inseminated from the
source and received by intermediate nodes (neighbors of the
source node). When RREQ is received by an transional node,
this fast check its roung table to nd a fresh route towards the
desnaon that is requested in RREQ. A route reply (RREP)
message is sent towards the source node through the pre-
established reverse route (established when RREQ pass through
intermediate nodes) if such a route is found. If the transional
node cannot able to nd a route, it restore of its roung table
and sends RREQ to its neighbors. This acon is repeated unl
the desnaon nodes receive the RREQ of source node [5].
Figure1 shows the Route Discovery procedure of Ad Hoc On-
Demand Distance Vector Roung where S is the source node and
D is the Desnaon node. Here A, C and B, E are the
intermediate nodes for traveling the RREQ message.
Figure 1: Route Discovery procedure of Ad Hoc On-Demand
Distance Vector Roung Protocol.
Every intermediate node in the network increases the hope
count one by one when the RREQ (Route Request) packet travels
through the network. If a Route Request (RREQ) message is
received in a node with the same RREQ ID or Broadcast ID, the
receiving intermediate node discard the newly received RREQs
with controlling the ID eld of the RREQ message. When the
RREQ and RREP messages are traveled through the network by
the transional nodes, the transional nodes update their
roung tables and save this route entry for 3 seconds. The
ACTIVE_ROUTE_TIMEOUT constant value of AODV protocol is
‘3seconds’ [6].
Black Hole Aack
Mobile Ad Hoc Network using the AODV protocol faces an
aack named Blackhole aack where a malicious node or
Blackhole node consumes the network trac and drops all data
packets. To explain the Black Hole Aack, an example is shown
in the following Figure2. In Figure 2, we assume that Node B is
the malicious node or Black hole node. When Node A broadcasts
the RREQ message for Node D to establish a path for data
transfer, Node B immediately responds to Node A with a false
RREP message showing that it has the highest sequence number
of Node D, as if it is coming from Node D. Node A assumes that
Node D is behind Node B with 1 hop count and discards the
newly received RREP packet come from Node C or E. Node A
then starts to send out all data packet to the node B. Node A is
trusng that these packets will reach Node D but Node B will
drop all data packets. The malicious node or Black hole node
takes all the routes coming up to itself. It stops forwarding any
packet to any other nodes. The network operaon is hampered
as the black hole node B consumes the packets easily [7].
American Journal of Computer Science and Information Technology
ISSN 2349-3917 Vol.6 No.2:25
2018
2This article is available from: https://www.imedpub.com/computer-science-and-information-technology/

Figure 2: A single Black hole aack in Mobile Ad Hoc Network
Implemenng a Roung Protocol in NS to
Simulate Black Hole Behaviour
To give a node the characteriscs of blackhole node we need
to implement a new roung protocol in ns 2.35. Implementaon
of a New MANET unicast Roung Protocol in NS-2 is described in
the reference. [8]. All roung protocols in Network simulator-
2.35 are installed in the directory of “ns-2.35”. We rst duplicate
the AODV protocol in the ns-2.35 directory and change the
name of this directory as “blackholeaodv”. In this blackholeaodv
directory the name of all les that are labeled as “aodv” are
changed to “blackholeaodv” such as blackholeaodv.cc,
blackholeaodv.h, blackholeaodv.tcl etc. All classes, funcons,
variables, and constants names in blackholeaodv directory have
changed but struct names that belong to AODV packet.h le
have not changed.
Figure 3: Adding the “blackholeaodv” protocol agent in the
“\tcl\lib\ns-lib.tcl” le.
To integrate the new blackholeaodv protocol in NS-2.35
simulator, we have changed two les that are used globally in
this simulator. In “\tcl\lib\ ns-lib.tcl” le we rst add the lines
shown in Figure-3, for the agent procedure for blackholeaodv.
Figure 4: Addion in the “\makele” at the ns-2.35 directory
Second le which is in the ns-2.35 directory named
“\makele” where we add the line shown in Figure 4.
In aodv.cc, the “recv” funcon process the packet based on
the type of the packet. If packet type is AODV route conducng
packet such as RREQ, RREP, RERR, it sends the packet to the
“recvAODV” funcon .When the received packet type is data
packet type then AODV protocol sends it to the desnaon
address. In the Figure 5 the rst “if” condion provides the node
to receive data packets if it is the desnaon and the “else”
condion consume all remaining packets as a Black Hole node.
Figure 5: “If” statement for accepng the packets by
desnaon or dropping packets by malicious node.
To generate the black hole behavior we need to make change
in blackholeaodv.cc le by adding the false RREP. The false RREP
message show that it has the highest sequence number and the
sequence number is set to 4294967295 and hop count is set to
1.The Highest sequence number of AODV protocol is
4294967295, 32 bit unsigned integer value [5]. The lines in
Figure 6 are added to aodv.cc le to generate the characteriscs
of black hole node. Aer changing the les then we compiled
the “make” in the terminal window (Cygwin window) to create
object les.
Figure 6: The false RREP of blackhole or malicious node.
Soluon for the Black Hole Aack on
AODV Protocol in Manet
To detect the blackhole aack the “Blackhole Detecon
System” checks the RREPs that come from mulple paths. As the
American Journal of Computer Science and Information Technology
ISSN 2349-3917 Vol.6 No.2:25
2018
© Under License of Creave Commons Aribuon 3.0 License 3

blackhole node immediately send RREP message to the source
without checking its roung table, it is more likely that the rst
RREP comes from the blackhole node. Then the soluon will
discard the rst RREP packet using the route reply saving
mechanism that come from malicious node and choose the
second RREP packet.
Algorithm for Black Hole Detecon
System
•Step1: Source node broad cast route Request (RREQ) packet.
•Step2: Mulple Route Reply of corresponding Route Request
comes to Source node.
•Step3: The Route Reply that comes rst set as the response
from malicious node and removes from the table by using
the RREP saving mechanism.
•Step4: The second Route Reply is choose by RREP saving
mechanism and set it as reply from corresponding
desnaon node. Then the source node delivers the data to
the path through which the second RREP came.
•Step5: Stop.
Implemenng the Black Hole Detecon
System in NS Against the Black Hole
Aack
To implement soluon against Blackhole, we duplicated the
“AODV” protocol, changing it to “bdsAODV” as we did in
“blackholeaodv”. Here for the soluon, we had to change the
receive RREP funcon (recv Reply) and create RREP saving
mechanism. This RREP saving mechanism counts the second
RREP message. At rst, we have changed all les name in the
cloned “aodv” directory to bdsAODV. To integrate the new bds
AODV protocol in NS-2.35 simulator, at First the le “\tcl\lib\ ns-
lib.tcl” is modied where protocol agents are coded that is
presented in Figure 7.
Figure 7: Adding the “proposed” protocol agent in the “\tcl
\lib\ ns-lib.tcl” le.
Second le which is in the ns-2.35 directory named
“\makele” where we add the lines that is in Figure 8. To detect
blackhole aack we create RREP saving mechanism in recv Reply
funcon of bdsAODV.cc le that is presented in Figure 9. In the
RREP saving mechanism the “rrep_insert” funcon is used for
adding RREP messages, “rrep_lookup” funcon is used for
looking any RREP message up if it is exist,“rrep_remove”
funcon removes any record for RREP message that arrived from
dened node and “rrep_purge” funcon is to delete periodically
from the list if it has expired.
Figure 8: Addion in the “\makele” at the ns-2.35 directory.
Figure 9: RREP saving mechanism in the bdsAODV Protocol.
We rst check if the RREP message arrived for itself, if it
arrived for itself then the funcon looks up RREP message if it
has soluon’s receive RREP message funcon is already arrived.
If it did not arrived then it inserts the RREP message for its
desnaon address and returns from the funcon. If the RREP
message is arrived or cached before for the same desnaon
address then the normal RREP funcon is carried out. If the
RREP message is not arrived for itself then the node forwards
the message to its appropriate neighbor. The code blocks
represented in Figure10 shown how the bdsAODV carried out.
American Journal of Computer Science and Information Technology
ISSN 2349-3917 Vol.6 No.2:25
2018
4This article is available from: https://www.imedpub.com/computer-science-and-information-technology/

Figure 10: Receive RREP funcon of the bdsAODV protocol.
Result Analysis
Here we simulate the same tcl script using AODV protocol
without black hole aack, AODV protocol with a single blackhole
aack, bdsAODV soluon with a single blackhole aack. We
simulated our model for 20, 25, 30, 35 and 40 nodes. Then we
compared the Performance Metrics such as the PDR (packet
Delivery Rao), End to End delay, through put and Jier of the
three scenarios.
Simulaon parameters
Table 1: Simulaon parameters
Parameter Definition
Protocol AODV, blackholeAODV, bdsAODV
MAC layer IEEE 802.11
Simulation area 700m*700m
Size of data packet 512 bytes
Traffic sources CBR/UDP
Number of nodes 20, 25, 30, 35, s40
Number of blackhole node 1
Antenna Type Antenna/Omni Antenna
Version NS 2.35
Performance evaluaon
We have been analyzed the result using four performance
metrics. They are the PDR (packet delivery Rao), End to End
Delay, throughput and Jier (Table 1).
Packet delivery rao (PDR): It is the rao of the number of
data packets received by the desnaon to the number of data
packets generated by the sources node. In the Figure 11, PDR
values for 20, 25, 30, 35 and 40 nodes for normal AODV,
blackholeaodv and bds AODV soluon are ploed. Here, it is
shown that PDR of AODV is aected by the malicious node and
the average packet delivery rao for this scenario is about
33.42%. Whereas the PDR of bdsAODV against Black Hole node
has high packet delivery rao compared to blackholeaodv and
it’s about 46.7%.
Figure 11: Packet Delivery Rao (PDR) vs. number of nodes.
End to end delay: The me taken by a packet to travel from
source to desnaon is called the End to End delay. In Figure 12,
the average end to end delay with single Black hole aack is
decreased in the bdsAODV soluon against Black hole aack.
Figure 12: Average End to End Delay vs. Number of nodes.
Throughput: Throughput can be dened as the amount of
data transferred from sender to receiver in a given amount of
me. It is measured in bits per second or packets per second. In
Figure 13, the bdsAODV soluon with single Black Hole aack
has average 5kbs-1 throughput which is greater than the
average throughput of blackholeaodv that is about 3.6 kb/s-1.
American Journal of Computer Science and Information Technology
ISSN 2349-3917 Vol.6 No.2:25
2018
© Under License of Creave Commons Aribuon 3.0 License 5

Figure 13: Throughput vs. number of nodes.
Jier: The variaon in the delay of received packet is called
jier. Low jier or Minimum variaon in the packet arrival me
provides beer performance in network. In blackhole scenario,
the average jier is about 12% whereas in bdsAODV soluon the
average jier is 7% that means it provide beer performance
presented in Figure14.
Figure14 : Average jier vs. number of nodes.
Conclusion
When we simulated the Black hole aack, we saw that the
data loss is occurred and the packet delivery rao was
decreased. Aer that, we simulated the bdsAODV soluon
against black hole aack and saw that the data loss that
occurred due to black hole node was decreased. The soluon
also increased the throughput and decreased the jier. PDR
(Packet Delivery Rao) of AODV is aected bythe Black Hole
node and the average packet delivery rao for this scenario is
about 33.42%. Whereas the PDR of bdsAODV soluon against
Black hole node has high packet delivery rao compared to
blackholeaodv and it’s about 46.7%. The soluon decrease the
average Jier 5% compared to the situaon of Black hole aack.
The advantage of this approach is that for implemenng the BDS
soluon we do not make any modicaon in packet format
hence can work together with AODV protocol. Another
advantage is that the soluon requires minimum modicaon in
AODV protocol.
References
1. Shrivastava P, Kumar S, Shrivastava M (2014) Study of Mobile Ad
hoc Networks. Int J Comp App 86: 0975-8887.
2. Kaur S, Gupta AK (2012) Posion Based Roung in Mobile Ad-hoc
Networks: An Overview. Int J Comp Sci Technol 3: 792-796.
3. Misra P (2006) Roung Protocols for Ad Hoc Mobile Wireless
Networks. hp://www.cse.wustl.edu/~jain/cis788-99/
adhoc_roung/index.html.
4. Kumar A, Reddy L, Hiremath P (2008) Performance Comparison of
Wireless Mobile Ad-Hoc Network Roung Protocols. Int J Comp Sci
Net Security 8.
5. Singh G, Singh J (2012) MANET: Issues and Behavior Analysis of
Roung Protocols. Int J Adv Res Comp Sci Soware Eng 2.
6. Perkins C (2003) (RFC) Request for Comments-3561. Category:
Experimental, Network, Working Group.
7. Deng H, Li W, Agrawal DP (2002) Roung Security in Wireless Ad
Hoc Networks. University of Cincinna, IEEE Communicaon
Magazine.
8. Ros FJ, Ruiz PM (2005) Implemenng a New Manet Unicast
Roung Protocol in NS2. hp://masimum.dif.um.es/nsrt-
howto/pdf/nsrthowto.pdf.
American Journal of Computer Science and Information Technology
ISSN 2349-3917 Vol.6 No.2:25
2018
6This article is available from: https://www.imedpub.com/computer-science-and-information-technology/