Recent Advances In Telecommunications, Informatics And Educational Technologies
Swarm based Energy Aware Mitigation Mechanism for Root Node
Attacks in MANETs
S.PARTHIBAN, 2PAUL RODRIGUES
Department of Computer Science Engineering
Anna University, Chennai
INDIA
1
[email protected], [email protected]
1
Abstract: - In Mobile ad hoc networks, the reliability in packet delivery during group communication highly
depends on the collaboration of intermediate nodes existing between the source and destination. Further, the
Rendezvous point node acts as the group leader for forwarding packets between in each and every multicast
group. Furthermore, the root node attack in multicasting may crumble the performance of the entire network.
Hence, the necessity for detecting and mitigating Rendezvous point is considered as one of the significant
research issue to be resolved. In this paper, we propose a Swarm based Energy Aware Mitigation Mechanism
(SEAMM) for mitigating rendezvous point attack based on Optimal Energy Factor (OEF) and swarm
intelligence. The performance of SEAMM is analyzed using ns-2 simulation and the proposed energy based
detection approach outsmarts the CONFIDANT protocol with an improvement in packet delivery rate and
throughput of 34% and 29% respectively.
Key-Words: -Rendezvous point attack, MAODV, Optimal energy factor, Mobility factor, Mobility Proportion,
Expected energy.
1 Introduction
Although, researchers have put forward large
number of techniques, to detect and mitigate various
types of malicious attacks in MANETs, most of the
proposed approaches were mainly framed for
unicast routing activity. The influence of malicious
attacks on the multicast application has not been
explored in the literature. This paper focuses on
detecting and mitigating malicious nodes in a
multicast routing activity by making use of
MAODV protocol. The MAODV protocol is a tree
based protocol, in which the data dissemination
from source group to destination group is done
through the rendezvous point present in each
multicast group. In this paper, we propose a Swarm
based Energy Aware Mitigation Mechanism
(SEAMM) for mitigating rendezvous point attack
based on Optimal Energy Factor and swarm. The
proposed approach SEAMM quantifies the trust of
the group leader is estimated through the Optimal
Energy Factor computed based on Observed Energy
and Expected Energy identified for a root node.
Further, new group leader that replaces the existing
group leader is elected based on constant mobility
factor.
Mobile ad hoc network is defined as a collection of
autonomous mobile nodes that performs the act of
routing without relying on any centralized
infrastructure. In this network, nodes are free to
move in an arbitrary fashion and hence the topology
of the network is highly dynamic in nature [1]. In
the dynamic topology, the mobile nodes present in a
particular range can communicate directly, whereas
the nodes present outside the communication range
make use of intermediate nodes to transfer a data
packet to its destiny and this type of transmission
may be called as multi-hop routing. In this multihop routing, the probability of a node participating
in a routing activity is highly dependent on the
reputation factor of the node [3]. The reputation
factor of a mobile node reflects the reliability and
cooperatively of the particular mobile node to
participate in a routing activity. But, there are some
classes of mobile nodes which do not actively
participate in the routing activity and drops many
packets without transmitting to the next intermediate
or to the destiny node. In general, such classes of
nodes are known as malicious nodes, which by its
activity drastically reduce the network performance.
ISBN: 978-1-61804-262-0
The remaining part of the paper is organized as
follows. Section 2 presents a brief literature review
199
Recent Advances In Telecommunications, Informatics And Educational Technologies
on the existing mitigation mechanism proposed for
detecting rendezvous point attack. Section 3 and 4
elaborates on the proposed SEAMM mechanism
with its associated algorithms. Section 5 presents the
simulation setup used for the implementation of
SEAMM and the comparative study carried out with
CONFIDANT. Section 6 concludes the paper with
future plan of research.
the network by utilizing a bootstrap router that
enables efficient routing. This mechanism is based
on the principle of tit for tat strategy where
malicious nodes are punished with -1 and the
normal routing nodes are rewarded with +1.
Depending upon the number of punishments and
rewards gained by a mobile node, the decision of
isolating malicious nodes were taken into account.
C. Demir and C.Comaniciu et al.,[9] proposed a
2 Related Work
novel detection mechanism based on the
concept of auction. This mechanism selects the
optimal routing path by computing the
minimum cost obtained by the mobile nodes in
each and every individual bids. This mechanism
further manipulates the optimal routing path
based on the payment which is equal to the
magnitude of second lower valued bid.
From the past decade, a number of mitigation
mechanisms have been proposed for rendezvous
point attack which is considered to be the crucial
malicious activity that could degrade the
performance of the network during multicasting.
Some of the mitigation approaches available in the
literature for detecting rendezvous point attack are
enumerated below.
In addition, H. Yang et al.,[10] contributed a
detection mechanism that utilizes one way hash
function for predicting the gentility parameter
of a mobile node. The utilized hash function
depends on the second hand information
obtained from the neighbour nodes of the
shared multicast group leader. They also
contributed a fault tolerance mechanism that
aids in enhancing delivery of data packets. An
evidence based trust aided mitigation
mechanism was proposed by Thomas M.Chen
et al., [11] which identifies the attacker nodes
based on the rule of Dempster-Shafer Theory.
This approach also computes the trust factor
based on second hand information using the
concept of posterior probability.
Kumar Das et al.,[4]contributed a reactive multicast
protocol that enables reliable transmission of data
by organizing nodes into shared multicasting
determined based on node behaviour. The authors
investigated the issue of multicasting based on link
stability, energy efficiency and expiration time.
They also developed a framework for efficient
detection of malicious activity that could decrease
the network survivability. L. Buttyan and J.P
Hubaux[5] contributed a trust oriented framework
that enforces cooperation between mobile nodes of
an ad hoc environment. They implemented a tamper
resistance module for detecting and mitigating
rendezvous point attack. They also prove that
malicious activity can be identified efficiently based
on path rater and watch dog mechanism.
Further, S.Roy et al., [6] contributed a mitigation
mechanism that could deal with the exhaustive sets
of attacks that could originate in a multicasting
environment.
This
mitigation
mechanism
investigates wide range of issues that could arise
during route establishment and route maintenance
by incorporating control packets like RREP-INV,
MACT (P), MACT (J) and MACT (J)-MTF.Bing
Wu et al., [7] proposed a Watchdog and Path Rater
dependent intrusion detection system that monitors
the trustworthiness of the mobile nodes participating
in group communication. Watch dog and Path Rater
were utilized for computing reputation of the mobile
nodes through which the decision of identification
may be incorporated.
2.1 Extract of the Literature
From the review of the literature from the research
works available for mitigating rendezvous point
attack, the following shortcomings were identified:
•
•
Furthermore, Chi-Yang Chang et al., [8] innovated a
mechanism for recovering compromised nodes of
ISBN: 978-1-61804-262-0
200
A swarm intelligence based mitigation
mechanism for root node attack based on energy
availability has not been proposed to the best of
my knowledge.
A distributed mitigation mechanism that detects
a group leader as compromised and elects a new
group leader based on mobility proportion has
not been much explored.
Recent Advances In Telecommunications, Informatics And Educational Technologies
These limitations of the literature are the motivation
factors for proposing Swarm based Energy Aware
Mitigation Mechanism for mitigating rendezvous
point attack based on Optimal Energy Factor and
swarm intelligence.
root node.
Then, Observed Energy
computed through (1)
(1)
2.1.1 Sub-subsection
When including a sub-subsection you must use, for
its heading, small letters, 11pt, left justified, bold,
Times New Roman as here.
The Expected Energy
computed through (2)
of a root node is
(2)
3 Swarm based Energy Aware
Mitigation Mechanism (SEAMM)
Then, Optimal Energy Factor
using (3)
is computed
(3)
In group communication, when a source node wants
to transfer data packets to the destination node
present in another group, the routing path from one
group to another group is identified based on the
group leader. In MAODV protocol which is a tree
based protocol, the group leader is a root node
present in the tree structure. When a root node or
group leader is comprised to perform malicious
activities like forwarding malicious node’s packets,
being idle without participating in the normal
routing activity, then such type of attack is termed
as root node attack.
If the obtained value of
is very less i.e., value
near to 0, shows that the root node does not
participate in any routing activity and also it shows
that the root node is compromised to the malicious
behaviour. Isolation process of the non-cooperative
root node is done and new group leader is selected
using the information obtained through swarm based
routing.
The election process of group leader through swarm
based routing is based on the following steps:
The proposed SEAMM approach detects the
presence of root node attack based on the
. This
computation of Optimal Energy Factor
factor quantifies the energy level of the root nodes
participating in a group communication by
considering the two parameters into account viz.,
1) The Observed Energy
of root node is
defined as the amount of energy utilized by a
mobile root node while participating in the
routing activity.
a) When a Source node in a multicast group wants
to communicate with a destination node of
another multicast group, it sends the RREQ tothe
Group leader (G1) of the sender at some level T1.
b) The Group leader (G1) at T1 initially searches the
local cache update which stores the energy level
of G1,for the matching process to compare the
energy level with all the downstream nodes.
c) If a downstream node of G1 contains energy
more than G1and further greater than half of the
energy available with the other downstream
nodes.
d) Then, GLEC triggers thedownstream node of G1
to be elected as the new G1by sending GRPH and
MACT packets all the other downstream and
upstream nodes.
e) On receiving the MACT packet, the Group leader
node G2will check their MACT_Table.
f) SR collects all the matching entries fromthe
MACT packets send by the Group leader node
G2.
g) Further, if a downstream node of G1 does not
contain energy more than G1and not greater than
half of the energy available with the other
downstream nodes
2) The Expected Energy
of a root node is
defined as the ratio of currents available in the
root node to the minimum energy required for
forwarding packets.
The Observed Energy
is computed based on
initial energy
energy utilization as follows. Let
possessed by the root node before the
be the currently
commencement of data transfer,
available energy of a root node,
be the energy
required for a root node to transfer a single data
packet and
be the number of packets present in
the root node buffer to be transferred to the another
ISBN: 978-1-61804-262-0
is
201
Recent Advances In Telecommunications, Informatics And Educational Technologies
h) GLEC selects a node with has a constant mobility
factor measured based on signal to noise ratio,
determined through bit error ratio (BER) given by
(4)
8. Else
9. Normal routing
10. End
When the source node wants to send some data to a
destination using MAODV protocol, the source
performs the route discovery process by flooding
RREQ packets to all the possible paths destined to
the destination through the forward route and
establishes an efficient route to the destination after
the acknowledgement of RREP packets. Further, the
source node present in one of the multicast group
relies on the group leader of that particular group
and the group leader of the other multicast groups
participating in the group communication. This
reliability of the group leader is estimated through
the Optimal Energy Factor computed (OEF) based
on Observed Energy for a root node and Expected
Energy for a root node obtained through step (3) and
(4) of algorithm1.If the Optimal Energy Factor
of a rendezvous point of a multicast group is
found to be less than 0.1 and converging to 0, then
the rendezvous point node is identified as
compromised and the new group leader is elected
using algorithm 2 using mobility factor, dynamic
distance between mobile nodes and mobility
Proportion
(4)
Where‘d’ is defined as the average of the dynamic
distance between two mobile nodes and ‘k’ is the
mobility proportion.
In ideal condition, i.e., when movement is almost
then,
constant,
(5)
4 Algorithm for Swarm based Energy
Aware Mitigation Mechanism
The algorithm for Swarm based Energy Aware
Mitigation Mechanism can be explained into two
phases, viz.,
a) Algorithm for the Computation of Optimal
Energy Factor
b) Algorithm for electing New Group Leader
4.1 Algorithm for the Computation of
Optimal Energy Factor
4.2 Algorithm for electing New Group
Leader
Notations
- Optimal Energy Factor
- Observed Energy
-Expected Energy
- Initial energy possessed by the root node before
the commencement of data transfer
- Currently available energy of a root node
- Energy required for a root node to transfer a
single data packet
- Number of packets present in the root node
buffer to be transferred to the another root node
Notations
SR – Source Node
-mobility factor
D-average of the dynamic distance between two
mobile nodes
k-Mobility Proportion
1. Begin
2. Send RREQ to the Group Leader
of the
sender at the level
3. for the entire network of downstream nodes
of
do
4. Compare energy level of with all other
downstream nodes, if
5. Then, Elect downstream of
as new group
leader
1. Begin
2. for the entire network do
3. Compute Observed Energy for a root node using
4. Compute Expected Energy
computed through
5. Using
Factor
and
of a root node is
5.1 Begin
5.2 For the entire downstream and upstream
nodes
5.3 Send GRPH and MACT Packets
5.4 End
, Compute Optimal Energy
through
, then
6. If
7. Call Elect Group Leader ()
ISBN: 978-1-61804-262-0
202
Recent Advances In Telecommunications, Informatics And Educational Technologies
Table 1: Simulation Setup
6. Update the MACT_TABLE of group leader
7.
sends MACT packets to SR
8. Else
8.1 Begin
8.2. Elect new group leader
based on
constant mobility factor computer
through
8.3 End
9. End
Algorithm 2 elects the new group leader by
comparing the energy level of group leader with
all other downstream nodes of that multicast tree,
and elects a new group leader
i.e,
and this group leader announces its succession to all
the downstream and upstream nodes of that
multicast group.But, when energy level of group
is greater than the other downstream
leader
nodes, i.e,
. Then, the new group
leader is elected based on constant mobility factor
manipulated through step (9) of algorithm 2.
5. Simulations and Experimental
Analysis
The comparative analysis of SEAMM with the
CONFIDANT protocol is exhaustively carried out
through simulation using network simulator of
version ns-2.26. Further, the proposed topology for
simulation contains 100 mobile nodes that
arbitrarily move around a terrain size of 1000x1000
with a constant bit rate of 60 packets/sec, channel
capacity of 2 Mbps, refresh interval time of 20
seconds and simulation time of 150 seconds
.Further, the SEAMM and CONFIDANT algorithms
are implemented with the same simulation setup
parameters and their performances are evaluated
based on packet delivery ratio, throughput, total
overhead and control overhead since ,the reliable
dissemination of data between the source and
destination depend mainly on the group leader of the
shared multicast tree. Furthermore, the rendezvous
point attack drastically reduces the packet delivery
rate and throughput which in turn increases the
number of retransmissions.
Parameter
Value
Description
No. of mobile
nodes
100
Simulation
Node
Type of
Protocol
MAODV
Multicast ad
hoc On
demand
Distance
Vector
Protocol
Type of
Traffic
40 packets per
Second
Constant bit
rate
Type of
Propagation
Two Ray
Ground
Radio
propagation
model
Simulation
Time
50m
Maximum
simulation
time.
Number of
packets used
1000
Maximum
number of
packets used
in simulation.
Channel
capacity
2 Mbps
Capacity of
the wireless
channel
5.1 Performance analysis for SEAMM
The performance of SEAMM is exhaustively
studied based on three experiments viz.
a) Experiment 1: Varying the number of mobile
nodes with number of root node attackers (s=5).
b) Experiment 2: Varying number of mobile nodes
with number of root node attackers (s=10).
c) Experiment 3: Varying the number of root node
attackers in increments of 10.
The following section enumerates on the
performance of the SEAMM when compared to
CONFIDANT protocol:
5.1.1
Experiment 1: Varying the number of mobile
nodes with number of root node attackers
Fig. 2 depicts the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
ISBN: 978-1-61804-262-0
203
Recent Advances In Telecommunications, Informatics And Educational Technologies
MAODV WITHOUT ATTACK with respect to
packet delivery ratio. Our proposed SEAMM
approach exhibits an improved performance in
packet delivery ratio from 17% to 21% and from
24% to 29% when compared to the benchmark
schemes like CONFIDANT and MAODV WITH
ATTACK. This improvement in the packet delivery
rate is achieved since SEAMM detects and mitigates
rendezvous point attack at a faster rate of 38%.
It is also obvious that our SEAMM approach
mitigates root node compromised nodes rapidly and
increases the throughput by reducing packet drops
in an average of 18%.
Fig. 4 portrays the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK based on total
overhead. Our proposed approach SEAMM exhibits
an optimal performance by reducing the total
overhead from 23% to 26% and from 29% to 39%
over CONFIDANT and MAODV WITH ATTACK.
90
85
Packet Delivery Ratio
80
75
70
16
MAODV
MAODV
MAODV
MAODV
65
15
60
50
10
MAODV
MAODV
MAODV
MAODV
15
14
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
20
25
30
35
No.of Mobile Nodes
40
45
Total Overhead
55
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
50
13
12
11
Fig. 2: Experiment 1-Performance Chart for SEAMM
based on Packet Delivery Ratio
10
10
It is also evident that SEAMM mitigates root node
attack at faster rate and improves the packet delivery
rate in an average of about 23%.
15
20
25
30
35
No.of Mobile Nodes
40
45
50
Fig. 4: Experiment 1-Performance Chart for SEAMM
based on Total Overhead
It is also proved that, SEAMM is identified as an
effective and efficient approach that minimizes the
total overhead by reducing the number of
retransmissions in an average of 24%.
Fig. 3 represents the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK based on
throughput. Our proposed SEAMM mitigation
approach demonstrates a phenomenal improvement
in throughput in par with the existing approaches
like CONFIDANT from 16% to 29% and from 23%
to 33% over MAODV WITH ATTACK. This
improvement in throughput is due to the efficient
energy mechanism incorporated for detecting root
node attackers that maliciously drops packets.
Fig. 5 represents the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK with respect to
control overhead. Furthur, the proposed swarm
based SEAMM approach gradually reduces the
control overhead from 21% to 16% and from 18%
to 15% than CONFIDANT and MAODV WITH
ATTACK.
85
4
80
1.5
x 10
1.4
70
65
60
10
MAODV
MAODV
MAODV
MAODV
15
Control Overhead
Throughput
1.45
75
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
20
25
30
35
No.of Mobile Nodes
MAODV
MAODV
MAODV
MAODV
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
1.35
1.3
1.25
1.2
40
45
50
1.15
1.1
Fig. 3: Experiment 1-Performance Chart for SEAMM
based on Throughput
ISBN: 978-1-61804-262-0
1.05
10
204
15
20
25
30
35
No.of Mobile Nodes
40
45
50
Recent Advances In Telecommunications, Informatics And Educational Technologies
Fig. 5: Experiment 1- Performance Chart for SEAMM
based on Control Overhead
85
It is clear that our proposed SEAMM mitigates the
root node attackers based on swarm intelligence and
reduces the control overhead in an average of 19%.
Experiment 2: Varying number of mobile nodes
with number of root node attackers(s=10)
Fig. 6 depicts the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK with respect to
packet delivery ratio. Our proposed SEAMM
approach exhibits an improved performance in packet
delivery ratio from 13% to 16% and from 18% to
22% when compared to the benchmark schemes
like CONFIDANT and MAODV WITH ATTACK.
This improvement in the packet delivery rate is
achieved since SEAMM detects and mitigates
rendezvous point attack at a faster rate of 38%.
75
70
65
60
10
MAODV
MAODV
MAODV
MAODV
15
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
20
25
30
35
40
45
50
Fig. 7: Experiment 2-Performance Chart for SEAMM
based on Throughput
It is also obvious that our SEAMM approach
mitigates root node compromised nodes rapidly and
increases the throughput by reducing packet drops
in an average of 15%.
Fig. 8 portrays the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK based on total
overhead. Our proposed approach SEAMM exhibits
an optimal performance by reducing the total
overhead from 23% to 26% and from 29% to 39%
over CONFIDANT and MAODV WITH ATTACK.
90
85
80
Packet Delivery Ratio
Throughput
5.1.2
80
75
70
65
60
MAODV
MAODV
MAODV
MAODV
55
50
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
14
13.5
45
13
15
20
35
30
25
No.of Mobile Nodes
40
45
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
12.5
50
Total Overhead
40
10
MAODV
MAODV
MAODV
MAODV
Fig. 6: Experiment 1-Performance Chart for SEAMM
based on Packet Delivery Ratio
12
11.5
11
10.5
10
It is also evident that SEAMM mitigates root node
attack at faster rate and improves the packet delivery
rate in an average of about 16 %.
9.5
10
20
35
30
25
No.of Mobile Nodes
40
45
50
Fig. 8: Experiment 2-Performance Chart for SEAMM
based on Total Overhead
Fig. 7 represents the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK based on
throughput. Our proposed SEAMM mitigation
approach demonstrates a phenomenal improvement
in throughput in par with the existing approaches
like CONFIDANT from 11% to 18% and from 20%
to 26% over MAODV WITH ATTACK. This
improvement in throughput is due to the efficient
energy mechanism incorporated for detecting root
node attackers that maliciously drops packets.
ISBN: 978-1-61804-262-0
15
It is also proved that, SEAMM is identified as an
effective and efficient approach that minimizes the
total overhead by reducing the number of
retransmissions in an average of 24%.
Fig. 9 represents the performance of SEAMM over
CONFIDANT, MAODV WITH ATTACK and
MAODV WITHOUT ATTACK with respect to
control overhead. Further, the proposed swarm
based SEAMM approach gradually reduces the
control overhead from 21% to 16% and from 18%
205
Recent Advances In Telecommunications, Informatics And Educational Technologies
to 15% than CONFIDANT and MAODV WITH
ATTACK.
x 10
1.14
1.12
MAODV
MAODV
MAODV
MAODV
WITH ATTACK
WITH SEAMM
WITHOUT ATTACK
WITH CONFIDANT
65
Throughput
1.16
1.1
Control Overhead
MAODV WITH SEAMM
MAODV WITH CONFIDANT
70
4
1.08
60
55
1.06
50
1.04
1.02
45
1
40
10
0.98
0.96
10
15
20
25
30
35
No.of Mobile Nodes
40
45
Fig. 9: Experiment 2- Performance Chart for SEAMM
based on Control Overhead
25
30
35
40
45
50
Fig. 12 depicts the performance of SEAMM over
CONFIDANT based on total overhead. The
proposed SEAMM approach exhibits a phenomenal
decrease in total overhead of about 23% to 31%
than CONFIDANT.
Experiment 3: Based on varying number
of root node attackers.
The performance of SEAMM is compared with the
existing reputation approach named CONFIDANT
based on packet delivery ratio is presented in fig.
10. Our proposed SEAMM approach exhibits a
phenomenal improvement in packet delivery ratio
than CONFIDANT from 14% to 19%.
12.5
12
Total Overhead
85
20
It is also clear that SEAM effectively and efficiently
detects root attacker nodes and mitigates them at
faster rate of 34% and further increases the
throughput to a maximum extent of 17%.
It is clear that our proposed SEAMM mitigates the
root node attackers based on swarm intelligence and
reduces the control overhead in an average of 16%.
5.1.3
15
Fig.11: Experiment 3- Performance Chart for SEAMM
based on Throughput
50
MAODV WITH SEAMM
MAODV WITH CONFIDANT
80
11.5
Packet Delivery Ratio
11
75
10.5
70
MAODV WITH SEAMM
MAODV WITH CONFIDANT
65
10
60
55
15
20
25
30
35
No.of Root Node Attackers
40
45
50
25
30
35
No.of Root Node Attackers
40
45
50
It is also clear that SEAM effectively and
efficiently detects root attacker nodes and mitigates
them at faster rate of 34% and further reduces the
total overhead to a maximum level of 23%.
Fig.10: Experiment 3- Performance Chart for SEAMM
based on packet delivery ratio
Further, it is also obvious that SEAMM increases
the packet delivery rate to a maximum extent of
21% than CONFIDANT.
The performance of SEAMM is compared with the
existing reputation approach named CONFIDANT
based on Throughput is presented in fig.11. Our
proposed SEAMM approach exhibits a phenomenal
improvement in throughput than CONFIDANT
from 24% to 29%.
ISBN: 978-1-61804-262-0
20
Fig.12: Experiment 3- Performance Chart for
SEAMM based on Total Overhead
50
10
15
206
Recent Advances In Telecommunications, Informatics And Educational Technologies
Fig. 13 depicts the performance of SEAMM over
CONFIDANT based on control overhead. The
proposed SEAMM approach exhibits a phenomenal
decrease in control overhead from 23% to19% than
CONFIDANT.
[3] Zouridaki,C, Mark,B.L, Hejmo,M and Thomas,R.K
(2005). ‘A quantitative trust establishment framework for
reliable data packet delivery in MANETs’, Proceedings of
the 3rd ACM Workshop on security of ad hoc and sensor
networks, vol 1, pp.1-10.
[4] Subir Kumar Das, B.S. Manoj, and C. Siva Ram
Murthy. Dynamic Core-Based Multicast Routing Protocol
for Ad Hoc Wireless Networks. In Proceedings of the
Third ACM International Symposium on Mobile Ad Hoc
Networking and Computing, pages 24–35, June 2002.pp
33-46
4
x 10
1.35
Control Overhead
1.3
1.25
[5] L.Buttyan and J-P, Hubaux, (2003), ‘Stimulating
Coperation in Self –organizing Mobile Ad hoc Networks”,
Mobile Computing and Networking’ pp 255-265.
1.2
1.15
10
[6] S. Roy, V.G. Addada, S. Setia and S.Jajodia, Securing
MAODV: Attacks and countermeasures, in Proceedings of.
SECON’05, IEEE, 2005.
MAODV WITH SEAMM
MAODV WITH CONFIDANT
1.1
15
20
35
30
25
No.of Root Node Attackers
40
45
50
[7] Bing Wu, Jianmin Chen, Jie Wu, Mihaela Cardei. A
Survey on Attacks and Counter measures in Mobile Ad
Hoc Networks. WIRELESS/MOBILE, NETWORK
SECURITY Y. Xiao, X. Shen, and D.-Z. Du (Eds.) 2006
Springer.
Figure 13: Experiment 3- Performance Chart for
SEAMM based on Control Overhead
It is also transparent that SEAM effectively and
efficiently detects root attacker nodes and mitigates
them at faster rate of 34% and further reduces the
control overhead to a maximum level of 16%.
[8] Chi-Yuan Chang, Yun-Sheng Yen. Chang-Wei Hsiesh
Han-Chieh Chao an Efficient Rendezvous Point Recovery
Mechanism in Multicasting Network, International
Conference on Communications and Mobile Computing,
2007.
6 Conclusion
[9] Demir, C and Comaniciu C. An Auction based AODV
Protocol for Mobile Ad Hoc Networks with Selfish Node.
Communications ICC'07. IEEE International Conference
in June 2007.
This paper has presented a Swarm based Energy
Aware Mitigation Mechanism (SEAMM) for
mitigating rendezvous point attack based on
Optimal Energy Factor (OEF) and swarm
intelligence. This mechanism identifies reliability of
the group leader through the Optimal Energy Factor
computed through Observed Energy and Expected
Energy possessed by the root node. The exhaustive
simulation study of SEAMM reveals that this
approach isolates rendezvous point attackers and
improves the performance of the network in terms
of Packet delivery ratio, Throughput, Control
overhead and Total overhead with a rapid detection
rate of 34% than the existing approaches of the
literature. As the part of our future plan, We have
proposed to elect the group leader based on link
stability metrics and link utility factor.
[10] H. Yang, H. Y. Luo, F. Ye, S. W. Lu, and L. Zhang,
Security in mobile ad hoc networks: Challenges and
solutions. IEEE Wireless Communications, vol. 11, pp. 3847, 2004.
11] Chen, T.M, Varatharajan,V (2005) Dempster-Shafer
Theory for Intrusion Detection in Ad Hoc Networks. IEEE
Internet Computing.pp 233-245.
References
[1] Rizvi,S and Elleithy,M,(2009) ‘A new scheme for
minimizing malicious behavior of mobile nodes in Mobile
Ad Hoc Networks’, IJCSIS Internation Journal of
computer Science and Information Security. Vol.3, No.1.
[2] Tarag Fahad and Robert Askwith., ‘A Node
Misbehaviour Detection Mechanism for Mobile Ad hoc
Networks’, PGNet. 2006.
ISBN: 978-1-61804-262-0
207