Secure Routing using Detection Method in Wireless Ad Hoc network

IJSRD - International Journal for Scientific Research & Development| Vol. 1, Issue 1, 2013 | ISSN (online): 2321-0613
All rights reserved by www.ijsrd.com
Ghada Vaseem J1
Prof. Sharada Valiveti2
1
M.tech(ICT) Student, Department Of Computer Engineering
2
Associate Professor , Department Of Computer Engineering
1,2
Nirma University, Ahmadabad, Gujarat
Abstract— An Ad Hoc network is the collection of multiple
nodes which can work together and they can send data over
multiple hopes without any infrastructure like base station
and antenna. Each node acts as a system and router. Many of
the routing protocols of Ad Hoc network are designed based
on the assumption that every node forwards every packet but
practically many of them act as selfish nodes, they use
network and its service but don’t cooperate with other nodes
so as to save resources for themselves. This paper discusses
the types of availability attack, malicious activity of selfish
node, a Survey of techniques used to detect selfishness
attack and some approach to detect selfishness attack.
Index Terms — Ad Hoc networks, Availability attacks, IDS
(credit based and reputation based techniques).
I. INTRODUCTION
An Ad Hoc network [1] is an infrastructure less network in
which many nodes are interacts with each other over
multiple hops without base station or access point. Network
topology is not maintained due to mobility of nodes. So
process of managing route information is of much
importance.
To enforce cooperation between nodes, trust is necessary.
Many of the routing protocols of Ad Hoc network are
designed based on the assumption that every node forwards
every packet. But practically many of them act as selfish
nodes. Sometimes, the node is not forwarding the packet
because they want to save their battery resources. Selfish
node can perform many malicious activities like remain in
sleep mode, does not forward route reply (RREP), does not
forward data packet, does not unicast or broadcast route
error (RERR), selectively drop packet etc.. Selfish node can
be handled either by punishing that node or by rewarding
that node which are not selfish. This paper presents the
survey of various selfishness attack detection scheme and
approaches for handling it.
II. AVAILABILITY ATTACKS
Availability is the most basic requirement of any network. If
the networks connection ports are Unreachable, or the data
routing and forwarding mechanisms are out of order, the
network would cease to exist [2]. Availability means that the
network should be in operational mode in the condition of
attack.
These are availability attacks:
1) Black hole attack [3]
Black hole attack is a denial of service attack in which a
malicious node can attract all the packets claiming a fresh
enough route to the destination and dropping all the packets
reaching at that node.
A black hole has two properties. (a) The node uses the Ad
Hoc routing protocol, such as AODV, to advertise itself as
having a valid route to a destination, even though the route
is fake, with the purpose of intercepting packets. (b) The
node consumes the intercepted packets. In an Ad Hoc
network that uses the AODV protocol, a black hole node
absorbs the network traffic and drops all packets.
2) Selfishness attack [3]
Cooperation among nodes in Ad-Hoc Networks is an
important issue for communication. But some nodes do not
cooperate in communication and saves their energy. These
nodes are called Selfish nodes. Selfish and malicious nodes
participate in route discovery stage properly to update their
routing tables, but as soon as data forwarding stage begins,
they discard data packets.
3) Resource consumption attack [3]
By sending unnecessary control messages, route request,
stale information, route discovery message, the malicious
node intentionally consumes the resources (Battery Power
and Bandwidth). In this attack, The Malicious node also
consumes Power and bandwidth by sending replication
message. Hence lifetime of network gets reduced.
4) Fabricated route attack [3]
Fabrication attacks generate false routing messages. Such
attacks can be difficult to confirm as invalid constructs,
especially in the case of fabricated false messages that claim
a neighbor cannot be contacted.
III. SELFISHNESS DETECTION
Basically there are two main techniques for preventing
selfishness attacks. These techniques fall under the category
of Trust based Routing. Selfish nodes are required to be
handled through their property of selfishness only. Hence
two methods are specified in literature:
1) Credit Based Schemes [4]
In this scheme for enforcing cooperation in performing the
network function faithfully, credits are provided to the node.
In the form of virtual currency based system is a node is
getting some fixed amount of credit for forwarding the
packet to next node and that node also pays some credit to
its next node for cooperation. To implement this approach
there are two models. 1. Packet Purse Model (PPM) and 2.
Packet Trade Model (PTM). In PPM, the sender who is
initiator pays some credit for getting packet forwarding
service, this packet forwarding charge is distributed among
Secure Routing using Detection Method in Wireless Ad Hoc network

(IJSRD/Vol. 1/Issue 2/2013/0060)
302
all forwarding nodes, initially the sender loads the packets
with some credits which are enough to reach destination. At
each intermediate node, packet should have some basic
credit to get service. If it doesn’t have enough credit then
that packet is discarded. The main problem with this
approach is to predict the total number of credits required
for whole transmission of packet from source to destination.
In PTM packet doesn’t carry the credit but the packet are
traded for credit by intermediate nodes. Each intermediate
node takes some credits from previous node and pays it to
next node. Total credit required for whole transmission is
covered by destination. The problem of PPM is solved here;
source node doesn’t require predicting the total number of
currency for service
2) Reputation Based Schemes [4]
In such schemes, reputation of each node of networks is
measured and based on this, it is decided whether the node
is malicious or not. After this the selfish node are discarded.
There are two models for reputation based schemes.
1. Watchdog Model [4]
2. Path rater [4]
Watchdog:
Figure 1 shows watchdog Model. As shown in Figure
1, suppose a path exists from node S to node D through
intermediate nodes A, B, and C. Here we can’t send packet
directly from node A to C. To send packet from S to D, A
sends packet to B and then node B sends it to C. This way
packet will reach to destination. Here each node maintains a
buffer for recently sent packets to check the match. If the
match is found, the packet is removed from buffer and
forgotten by the watchdog, since it has been forwarded on.
If a packet has remained in the buffer for longer time, the
watchdog considers it as failure and increments a failure
counter for the node responsible. If the tally exceeds a
certain threshold bandwidth, it means that the node is selfish
and about it the notification is sent to source node. The
problems with watchdog are power limitation, collision at
receiver, ambiguous collision and partial dropping.
Figure -1 Watchdog scheme [4]
Path rater:
To check reliability of each path in the network,
each node is preloaded with path rater. It gives the rate to
path by averaging the reputation of each node of that path.
If there are multiple paths to reach destination in network,
the path which has highest rate is selected for transmission
of packet. Table 1 shows the various methods for detecting
selfish nodes in Ad Hoc networks
Scheme Description Advanta
ge
Disadvant
age
End-to-end
Acknowledg
Monitoring
the reliability
Avoid
sending
Lack of
Misbehavi
ments [5] of route by
end to end
acknowledgm
ents
packets
through
unreliable
route
ng
detection
ABO(Activit
y based
overhearing)
[5]
Generalizatio
n of watchdog
Increase
the
number
of
observati
on and
improve
watch
dog
efficiency
. It
mitigate
collision
problem.
_
Two hop
acknowledg
ments. [5]
It uses
asymmetric
cryptography
Mitigate
power
control
technique
usage
problem
of
watchdog
_
Probing [5] Combination
of route and
node
monitoring
Probe
packet
for
selected
node to
detect
misbehav
ior
When
selfish get
probe
packet, it
choose to
cooperate
and
forward
packets for
limited
time.
CORE(colla
borative
reputation)
[5]
It can be
applied to
both data and
request
packets
forwarding
function
Watchdo
g for
monitorin
g both
directed
and
broadcast
ed
packets.
Watchdog’
s
drawbacks
are present.
CONFIDAN
T [5]
Cooperation
of nodes and
fairness in
dynamic ad
hoc network
Four
compone
nts in
each
node.
-
Watchdo
g monitor
-Trust
-
Watchdog’
s problem
-
Reputation
exchange
process can
cause
overhead

303
manager
-
Reputatio
n system
-Path
manager
Friends and
foes [5]
Set of friends
and foes
-it is used
to secure
control
packet
from
dropping
-watchdog
‘s problem
-More
overhead
-Mobile
selfish
node
cannot
detect
Ex-
Watchdog
[3]
- Maintaining
table
(source,dest,to
tal no. of
packets)
-detection of
malicious
node which
can partition
network
Solve
problem
of
watchdog
Fails when
malicious
node is on
all path
Table -1 Study of Techniques for detection of selfish node
IV. PROPOSED APPROACH
Here we propose two approaches to enforce cooperation
with Selfish Nodes:
A. Approach – I
In first approach we will use reputation based system to
detect selfish node. This approach helps to detect selfish
node, encourages and enforces the selfish node to cooperate.
This approach works in three phases to detect selfishness. In
Phase one, Monitoring and observing the behavior of one
hop nodes is done by each node using its own watchdog
module. And then it keeps the information about the number
of packets sent and received by each node. After every
specific time interval, watchdog upgrades the saved
information. In second phase, a cooperation coefficient A is
calculated which is same as the reputation of that node. This
reputation coefficient of any node is measured by taking the
proportion of Number of sent packets and Number of
received packets. This coefficient is a number whose value
may be between 0 and 1. The low coefficient value of node
which is near to zero indicates that the node is selfish node
and the high value of coefficient which is near to one
indicates the high cooperation of that node. After calculating
the reputation one new additional field is added to the
header of routing protocol. Here each node has the table for
maintaining the reputation. In phase three, priority of node
for processing packets is decided. The node whose
reputation value is high ,has the higher priority for getting
network services like sending packets while the node which
has low reputation value will get low priority for getting
services.
Figure-2. Flow diagram for reputation coefficient based
mechanism.
B. Approach – II
In second approach we will use credit based system. It
will encourage and enforce the selfish nodes to cooperate. In
this approach virtual currency system will be setup for node
to perform networking operation faithfully. Here each
intermediate node is paid for forwarding packets of another
node and the originator have to pay them. In this approach,
initially each node is preloaded with some fixed credit. For
every communication a fixed amount of virtual currency
would be deducted from the sender’s account and will be
distributed evenly distributed among the intermediate nodes.
Here the selfish node are not punished but the intermediate
nodes are rewarded the currency for cooperation. The
drawback of this approach is that the selfish node can attract
more routes through it by flooding wrong message of strong
routes and after this it can perform malicious activity on the
packets.
Figure-3 Flow Chart for Credit based mechanism.

304
V. PACKET DROPPING ATTACK
Step-1
Define a node as malicious node in tcl file
$ns- at 0.0 “[$node-(0) set ragent-] hacker”
$ns- at 0.0 “$node-(0) label ”Malicious Node””
Step-2
Now check the index of malicious node and
packets type weather it is same as tcl file or not and the
packet is data packet or not. If the condition is satisfied than
make the ttl (time to live) of data packets as zero and then
drop the packets by calling drop () function and increase the
drop index value by one. Now declare as selfish node when
packet drop index value cross the some maximum threshold.
Static into d [5000];
Void forward ()
{ if (index==4 && chptype()!=PT_AODV)
{ ih ttl=0; }
If (ihttl==o)
{ d[index] = d[index] +1;
Drop(p,DROP-RTR-TTL);
}
If (d[index] > 25)
{
Printf (“index= %d” is malicious node,index);
} }
VI. SIMULATION & RESULT
PROPERTY VALUE
Simulator NS-2.34
Nodes 25
Simulation Time 600
Mobility Model Random Way Point
Coverage area 1000m * 1000m
Mobility speed 10m/s
Pause time 2.0 sec
Connections 15
Traffic Types Constant Bit Rate
Packet size 512 bytes
Table -2 Simulation Parameters
Figure -4 Throughput Comparisons
Figure-5 Latency Comparison
Figure -6 PDR Comparisons
VII. CONCLUSION
In this paper we have discussed types of availability attack
in wireless Ad Hoc network. This paper presents the role
and malicious activity of selfish node in network and a
survey of intrusion detection system for selfishness attack in
Ad Hoc network. Here Reputation and Credit based, two
main detection techniques and their module are discussed in
detail. In addition two algorithms for selfishness detection
are proposed, one of them credit based and another one is
reputation based system. In reputation based approach the
selfish nodes are punished for their malicious activity and
other nodes are getting batter priority for their reputation
where in credit based approach instead of punishing the
malicious node, the non-malicious node are rewarded for
their cooperation in network functionality. Here from the
result we can conclude that in the presence of malicious
node Packet Delivery Ratio and Throughput decrease. With
the help of the proposed approach we can improve that both
the factor.
REFRENCES
[1] A.P. Yih-chun Hu, David B. Johnson, “Sead: Secure
efficient distance vector routing for mobile wireless Ad
Hoc, networks,”2003.
[2] A.Burg,” Ad hoc network specific attacks,”2003.
[3] Niyati Shah, Sharada Valiveti,”Intrusion Detection
System for the Availability Attacks in Ad-Hoc
Networks,”2011.

305
[4] Djamel Djenouri, Nadjib Badache,”MANET: Selfish
Behavior on Packet Forwarding”, Encyclopedia of
wireless and Mobile communication, 2008
[5] Marti,Giuli,Lai,Baker,”Mitigating routing misbehavior
in mobile adhoc networks”,ACM conference on mobile
computing and networking,2000
[6] Kargl,Klenk,Weber,Schlott,”Advance detection of
selfish or malicious node in adhoc networks”, 1st
European workshop on security in adhoc and sensor
networks,2004.
[7] Djenouri, Badache,” A novel approach for selfish node
detection in MANETs: proposel and petri nets based
modeling”,8th
IEEE internation conference on
telecommunication,2005
[8] Awerbuch,Holmer,Nita,Rubens,”An on demand secure
routing protocol resilient to byzantine failure”,ACM
workshop on wireless security,2002
[9] Michiardi, Molva, ”CORE: collaborative reputation
mechanism to enforce node cooperation in MANET”,
6th IFIP communication and multimedia security
conference, 2002

Secure Routing using Detection Method in Wireless Ad Hoc network

More Related Content

What's hot (17)

Viewers also liked (18)

Similar to Secure Routing using Detection Method in Wireless Ad Hoc network (20)

More from ijsrd.com (20)

Recently uploaded (20)

Secure Routing using Detection Method in Wireless Ad Hoc network