Detecting Misbehaving and Selfish Nodes in the Network using Watchdog Mechanism

International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016]
Infogain Publication (Infogainpublication.com) ISSN: 2454-1311
www.ijaems.com Page | 404
Detecting Misbehaving and Selfish Nodes in the
Network using Watchdog Mechanism
K. Naveen Kumar1
, Assistant Prof. K. Sriprasadh2
1
P.G student, Thirumalai Engineering College, kilambi, kanchipuram, Tamil Nadu, India.
2
Asst Prof. CSE, Thirumalai Engineering College, kilambi, kanchipuram, Tamil Nadu, India.
Abstract— The nodes in a wireless network may misbehave at
times. This misbehavior needs to be monitored in order to
avoid sudden failure of network. The watch dog mechanism
has been sufficiently studied to address the issue of malice
node detection, in Mobile Adhoc Networks (MANETs). A
Collaborative Contact based Watchdog (CoCoWa) is
collaborated with information diffusion in the proposed work.
This combination strategy analyses all the nodes in a network
and provides the information update regarding the selfishness
of the specific nodes to other nodes and routing protocols to
enable performance oriented transmission. Once the selfish
node is detected by the watch dog, it is marked as selfishness
positive node else the node is marked as negative selfish node.
For enabling this fool proof approach, true neighbors, fake
neighbors, their probability of relationships with each other is
analyzed. The evaluation of the viability of the proposed work
is made in terms of detection efficiency, detection accuracy of
both malicious and selfish nodes. Apart from these, the
strategy is proved to be simple yet effective.
Keywords— Wireless Networks, MANET, Selfish Nodes,
Watch Dog, Information Diffusion, Performance analysis.
I. INTRODUCTION
In the past few decades, Mobile Adhoc Networks (MANET)
has been widely addressed because of high usage of mobile
devices. To be specific, the MANET is the set of mobile nodes
setup in a wireless multihop network without a determined
infrastructure. This type of network is dynamic in nature that
are totally independent from fixed organizations.
It has the following features
• Self-created
• Self-organized
• Highly dynamic
• Capable to reconfigure
In this environment, the mobile nodes are moving randomly
without any centralized administration. The message
transmission between two distant nodes can happen only by
relay through the intermediate nodes in the network.
Fig.1: Selfish nodes
Fig 1 describes the working of selfish nodes in the network.
Node 3 and node 5 is the selfish nodes as they don’t forward
the packets they receive. The threats associated with selfish
nodes can be dangerous to the trust of the concerned network
and security of data. When the data is transferred from one
mobile node to another node in a particular network, someone
may hack or perform some malicious activities in the network.
During the information transfer or when the services are
offered to a user, the user may face delay reception of data.
The information delivery to the destination node becomes
delayed. These scenarios are considered as malicious
activities or selfish activities. In DTNs, selﬁsh nodes can
seriously degrade the performance of packet transmission. For
example, in two-hop relay schemes, if a packet is transmitted
to a selﬁsh node, the packet is not retransmitted, therefore
being lost. Thus the co-operation of nodes is mandatory to
work in the MANETs. The network design strategy adopted
can be modified to accomplish this, however the precautions
to know the presence of selfish nodes also becomes necessary.
This co-operation can be described in terms of the contact. The
contact is the relationship that a node have with the other
nodes during the presence of communication range. Tough it
has been known that, node co-operation is expensive, it has to
be attained to sustain the reputation or trust of the network. In
the earlier works, it has been provided with
• Incentive based methods and

• Detection and Exclusion strategies
For these enable stimulation of nodes in the network to
actively participate in the routing of packet forwarding
process. And our work does not focus to isolate or exclude the
selfish node but only to detect them positively without false
alarms.
This paper proposes CoCoWa as a collaborative contact-based
watchdog to reduce the time delay in packet transmission
through the nodes of the network and to improve the selfish
nodes detection efficiency. Here, the harmful effects of false
positives, false negatives and malicious nodes are reduced.
The proposed work combines the CoCoWa mechanism with
the diffusion of the known positive and negative detections.
The work focusses to obtain the less detection time (and
overhead) of a selfish node in a network.
The advantages of the proposed system are
• It takes less detection time (and overhead) of a selfish
node in a network.
• It reduce the time and improve the effectiveness of
detecting selfish nodes, reducing the harmful effect of
false positives, false negatives and malicious nodes.
• If the collaboration approach was not used, CoCoWa can
reduce the overall detection time with respect to the
original detection time
• Reduced overhead (message cost).
The following part of the paper is organized as follows.
The prior works related to the selfish node detection
mechanisms are discussed with their methodologies and
challenges associated with the anomalies detection and
information updates of those systems in section II. Section III
gives a detailed information on the proposed Collaborative
Contact based Watchdog (CoCoWa) with the diffusion
strategy and their working. Section IV gives the evaluation
which is supported by the simulation screen shots. The
concluding words are provided in section V.
II. RELATED WORK
Hernandez-Orallo, et al [1] suggested an analytical prototype
to find the selfish nodes with the help of watch dog in the
Mobile Adhoc Network (MANET). The model used contact
dissemination over the identified selfish nodes collaboratively
with the watch dog. Their metrics were evaluated in terms of
detection time and cost of implementation. Also in the next
work [2] studied the selfish nodes using watch dog with
respect to diffusion. Their metrics were evaluated in terms of
diffusion time and overhead of implementation. The work
focused on the false positives and false negatives. Li, et al [3]
evaluated the selfish behavior of the nodes of Delay Tolerant
Networks (DTN). The work evaluated the factors affected due
to the presence of selfish node. 3-D continuous time markov
chain based message delivery was undertaken. The study
concluded that the size of the multicast also decide the
performance of the system. Mahmoud and Shen [4] proposed a
secure cooperation incentive protocol for avoiding the
disruption due to riding attacks, to stop the denial of payment
and to minimize the overhead. For this purpose, public key
operations and light weight hashing operations were utilized.
Along with the performed operations, hash chains and keyed
hash values were applied. The results presented had reduced
overhead during transmission and proper accuracy.
Passarella and Conti [5] presented an analytical model to
describe the relation between individuals and aggregates
assuming a heterogeneous network. The aggregate inter-
contact time statistics was determined. Serrat-Olmos, et al [6]
proposed an integrated approach to identify the black holes
and selfish nodes in the MANET. A set of watchdogs were
utilized for the improvement of individual as well as collective
performance of the network. There was minimum time
requirement to detect the selfish node with maximum
accuracy. Gupta, et al [7] summarized the impacts of selfish
nodes in a Mobile Adhoc Network. The presence of selfish
node in a network would degrade the efficiency of the network
and ultimately the whole network would fail stated by the
proposed work. The concentration of selfish nodes were
illustrated along with the factors influenced by the nodes.
Padiya, et al [8] reviewed on the novel and existing
methodologies for the detection of selfish nodes in the
MANET. The work gave a scope for new findings for the
existing challenges. Han, et al [9] studied different trusted
approaches of Wireless Sensor Networks (WSN) in detail. The
work had compared and analyzed various parameters of the
study and presented the requirement for a robust approach for
the WSNs. Dias, et al [10] presented a novel co-operative
watch dog based system for detecting the misbehaving nodes
in Vehicular delay Tolerant Networks (VDTN). Thus the
identification was carried out for increasing the overall
network performance, probability of bundle delivery. The
resources were effectively utilized. Ataie, et al [11] suggested
a new idea to deal with the problem of selfish nodes in mobile
Adhoc networks. A Cooperation Enforcement, Malice
Detection and Energy Efficient Mechanism for Mobile Adhoc
Network (CEMDEEM) was introduced for identification and
isolation of energy based classical selfish nodes as well as
malice behaving nodes. The work investigated the network by
introducing a group of selfish nodes. The Dynamic Source
Routing (DSR) Protocol was executed and the performance
results were recorded. The work showed optimal performance
in the presence of up to 40 selfish nodes. Wahab, et al [12]

proposed a two phase model for treating the misbehaving node
in Vehicular Adhoc Network (VANET). The model utilized
Quality of Service Optimized Link State Routing (QoS-OLSR)
Protocol. The work focused on
• Identification of misbehaving vehicles that crossed the
speed range stipulated
• Promoting co-operation of selfish nodes during cluster
formation
• Detection of malice nodes after cluster formation
• Dempster- Shafer was used for the co-operative watch
dog model
The concept proved good for VANET with decreased false
negatives and increased detection probability. Also the Quality
of Service and stability was also maintained.
Sengathir, et al [13] reviewed and categorized the existing
solution proposed in various works for selfish node detection.
The misbehaviour was found to be the result of current
condition instead of its future or past history. The study
elaborated on the benefits and shortcomings of the markovian
models. Nikmaram [14] worked out on the following
objectives
• Evaluation of misbehaving node detection using current
acknowledgement methodology.
• Design and development of misbehaving node detection
using the acknowledgement strategy
• Provided energy efficiency
• Performance analysis of the proposed methodology
Akhtar and Sahoo [15] proposed an innovative mathematical
model for the classification of nodes based on their
misbehavior. Each node was graded based on the level of
misbehavior. The degree of tolerance towards misbehavior
was predefined so that, the nodes that misbehave was easily
identified. The messages was displayed as a course of action
against misbehavior. Thus, after effective classification of
misbehaving nodes, the list of nodes prone to misbehave were
sent to the routing protocol to avoid choking of data.
However the demerits of the existing system was listed as
follows
• There is no selfishness prevention mechanism is present.
• The packet delivery rates become seriously degraded.
• The number of packet losses is increased by 500 percent
when the selfish node ratio increases from 0 to 40 percent.
III. PROPOSED METHOD
In the proposed system, a new scheme for detecting selfish
nodes is established as Collaborative Contact based Watchdog
(CoCoWa) integrated with the information diffusion
mechanism. It combines local watchdog detections and the
dissemination of this information to the whole network. A
common technique to detect this selfish behavior of a node is
to monitor the network using local watchdogs. These watch
dogs points out the selfish and misbehaving nodes and the
information on the list of selfish nodes are diffused to all the
other nodes in the network to beware before transferring data
to that node. The information is updated regarding the status of
the nodes in the whole network. Thus with the help of all these
modules, the selfish nodes are detected successfully.
Fig.2:Generalized architecture of the proposed System
The modules of the proposed system is divided into five sub
modules such as
• Network Creation
• Packet Transmission
• Local Watchdog
• Diffusion
• Information Update
• Selfish node Prediction
Fig.3: Proposed Flow of Watchdog Mechanism

The modules are described in detail
1. Network Creation:
Network creation is first and fore most step for wireless
communication. Here, in this work the network nodes are
deployed over an area dynamically. The source node and
destination nodes are determined in this module.
2. Packet Transmission:
After the node deployment in the network the packet or file is
chosen to transfer from the source node to destination node.
Since the transmission is through wireless networks, it carried
over to the destination via a chosen communication channel.
3. Local Watchdog:
Watch dog proposed in the current work has the duty to
monitor the presence of selfish node and has to report
accordingly. Each node is assigned with the local watch dog,
which can overhear the packets transceived by its neighbors in
order to detect anomalies, such as the ratio between packets
received to packets being retransmitted. By using this
technique, the local watchdog can generate a positive (or
negative) detection in case the node is acting selfish (or not). It
is based on the combination of a local watchdog and the
diffusion of information when contacts between pairs of nodes
occurs. A contact are defined as an opportunity of transmission
between a pair of nodes. The Local Watchdog of the proposed
work has two functions: the detection of selfish nodes and the
detection of misbehaving nodes. In order to perform its
function effectively, they are designed to generate event based
on the occurrence of the selfish nodes. The relationship, a node
have with other normal nodes and misbehaving nodes is vital
to identify the selfish nodes in a MANET. This with ease the
work of proper routing possibility without channelizing the
data to the selfish nodes. The local watchdog can generate the
following events with regard to the neighboring nodes:
• PosEvt (positive event): when the watchdog detects a
selfish node, it gets initiated
• NegEvt (negative event) : when the watch dog detects
that a node is not selfish, this event is generated
• NoDetEvt (no detection event) when the watchdog
does not have enough information about a node.
4. Diffusion:
A key issue of our approach is the diffusion of information.
This is the novelty of the proposed work to be appended with
the CoCoWa approach. The Diffusion module has two
functions such that
(i) The transmission as well as the reception of information
regarding positive (and negative) detections.
(ii) The routing protocol is also kept informed about the
malicious nodes in the network to avoid forwarding data to
those nodes.
As the number of selfish nodes is low compared to the total
number of nodes, positive detections can always be
transmitted with a low overhead.
5. Information Update:
Updating or consolidating the information is the vital need of
the work to serve the purpose of the objective. A node can
have the following internal information about other nodes: No
Info state, Positive state and Negative state.
• No Info state : means that the module has no information
about that particular node,
• Positive state: denotes that the module believes that that
particular node is selfish,
• Negative state: means that the module believes that that
particular node is not selfish.
6. Selfish node Prediction
Finally, after the identification, diffusion and updation of the
malicious node in a network, the system can easily predict the
selfish nodes by based on
• the node behaviors in the network
• states of the nodes in the network
• events of the nodes in the network
IV. PERFORMANCE ANALYSIS
The performance evaluation has been done by simulating two
or three machines process in real-time means over an virtual
environment. Here we use machines as node instead of mobile
phones. Evaluation of the selfish node detection through the
wide nodes in the network enables the improved node specific
recognition process. The network created is evaluated with the
CoCoWa and diffusion modules. The proposed system is
evaluated with the Java coding run over MySQL database. The
efficiency of the proposed system is evaluated based on the
following metrics namely
• Accuracy
• Computational Complexity
• Detection Time
• Malicious Node detection efficiency
• Node Event Detection Accuracy
Let us discuss these parameters with the help of graphical
illustrations.
1. Accuracy:
Accuracy is defined as the ratio of the sum of true positive and
true negative values to the total number of frequent patterns.
True positive value denotes the number of correct frequent
patterns that are identified and true negative value denotes the
number of incorrect frequent patterns. Mathematically, the
accuracy is given by the following equation.

Accuracy =
∑ TP + ∑ TN
∑ Total frequent patterns
Fig.4 provides the comparative results of the existing and
proposed CoCoWa based diffusion method. Thus the graph
shows the improved accuracy in detecting the selfish nodes in
the network.
Fig.4: Accuracy analysis of the proposed system with the
existing System
2. Computational Complexity:
The computations involved to attain the purpose , that is to
detect the selfish nodes present in the network is found to be
less. Hence the complexity of the system is low. This simple
scheme of evaluation of the MANET in terms of misbehavior
is graphically illustrated in Fig. 5. Thus our system
outperforms the existing systems with reduced computational
complexity.
Fig.5 Computational complexity analysis of the proposed
system with the existing System
3. Detection Time:
The execution time is defined as the time spent by the
proposed CoCoWa based diffusion for finding the selfish node
distributed in the wireless networks. The execution time of the
proposed CoCoWa based diffusion is compared with the
execution time of the existing methods. Figure.6 shows the
execution time analysis of the proposed CoCoWa based
diffusion and existing methods. From the comparison results,
it is clearly evident that the execution time of the proposed
CoCoWa based diffusion is lower than the execution time of
the existing methods.
Fig.6: Detection time comparison of the proposed system with
the existing System
4. Malicious Node Detection Efficiency:
The effectiveness at which the proposed strategy detects the
malicious node in the wireless network is known as the
efficiency of malicious node detection. Fig. 7 provides the
analysis of the efficiency of CoCoWa based diffusion method
and that of the existing methods.
Fig.7: Malicious Node Detection time comparison efficiency
0
10
20
30
40
50
60
70
80
90
1 2 3 4 5
Accuracy(%)
Selfish Node Detection
Proposed System
Existing System
0
10
20
30
40
50
60
70
80
90
Existing System Proposed System
Complexity(%)
Computational Complexity
66
68
70
72
74
76
78
80
82
84
86
Existing System Proposed System
Milliseconds
Time Taken
0
10
20
30
40
50
60
70
80
1 2 3 4 5
Efficiency(%)
Malicious Node Detection
Proposed System
Existing System

5. Node Event Detection Accuracy:
The false alarms given by the proposed solution is negligible.
Thus the accuracy of detecting the node event is optimally
good rather than the existing approaches which may creace
false detection, which may result in disruption of network. Fig.
8 depicts the comparative analysis of the existing and
proposed CoCoWa based diffusion method.
Fig.8: Node Event Detection accuracy comparison of the
proposed system with the existing System
V. CONCLUSION AND FUTURE WORK
In the proposed work, the relationship or contact that a node
have with the other neighboring nodes are exploited to address
the issue of selfish nodes in the network. Each node is
assigned with the local watch dog to monitor and report their
misbehavior against the packet transmission in the network.
The information diffusion strategy is also integrated to give
the benefit of information update to the whole network to
avoid delay in packet transmission with the blame of the
presence of selfish nodes. Based on the event identification
and node status updation, the selfish nodes can be predicted.
This methodology is proved to outperform the earlier
approaches with the simulations performed to evaluate the
performance. Thus, the proposed system have greater selfish
node detection accuracy, node event detection accuracy,
malicious node detection. Also, it takes less time for detection
with the help of this new approach with less computational
complexity.
REFERENCES
[1] E. Hernandez-Orallo, M. D. Serrat, J.-C. Cano, C. T.
Calafate, and P. Manzoni, "Improving selfish node
detection in MANETs using a collaborative watchdog,"
IEEE Communications Letters, vol. 16, pp. 642-645,
2012.
[2] E. Hernández-Orallo, M. D. Serrat Olmos, J.-C. Cano, C.
T. Calafate, and P. Manzoni, "Evaluation of collaborative
selfish node detection in MANETS and DTNs," in
Proceedings of the 15th ACM international conference
on Modeling, analysis and simulation of wireless and
mobile systems, 2012, pp. 159-166.
[3] Y. Li, G. Su, D. O. Wu, D. Jin, L. Su, and L. Zeng, "The
impact of node selfishness on multicasting in delay
tolerant networks," IEEE Transactions on Vehicular
Technology, vol. 60, pp. 2224-2238, 2011.
[4] M. E. Mahmoud and X. S. Shen, "Esip: Secure incentive
protocol with limited use of public-key cryptography for
multihop wireless networks," IEEE Transactions on
Mobile Computing, vol. 10, pp. 997-1010, 2011.
[5] A. Passarella and M. Conti, "Characterising aggregate
inter-contact times in heterogeneous opportunistic
networks," in NETWORKING 2011, ed: Springer, 2011,
pp. 301-313.
[6] M. D. Serrat-Olmos, E. Hernández-Orallo, J.-C. Cano, C.
T. Calafate, and P. Manzoni, "A collaborative bayesian
watchdog for detecting black holes in MANETs," in
Intelligent Distributed Computing VI, ed: Springer, 2013,
pp. 221-230.
[7] S. Gupta, C. Nagpal, and C. Singla, "Impact of selfish
node concentration in manets," International Journal of
Wireless & Mobile Networks, vol. 3, 2011.
[8] S. Padiya, R. Pandit, and S. Patel, "Survey of innovated
techniques to detect selfish nodes in MANET,"
International Journal of Computer Networking, vol. 3,
pp. 221-30, 2013.
[9] G. Han, J. Jiang, L. Shu, J. Niu, and H.-C. Chao,
"Management and applications of trust in Wireless
Sensor Networks: A survey," Journal of Computer and
System Sciences, vol. 80, pp. 602-617, 2014.
[10]J. A. Dias, J. J. Rodrigues, F. Xia, and C. X.
Mavromoustakis, "A cooperative watchdog system to
detect misbehavior nodes in vehicular delay-tolerant
networks," IEEE Transactions on Industrial Electronics,
vol. 62, pp. 7929-7937, 2015.
[11]E. Ataie, A. Movaghar, and M. Bastam, "A Cooperation
Enforcement, Malice Detection and Energy Efficient
Mechanism for Mobile Ad hoc Networks," International
Journal of Sensors Wireless Communications and
Control, vol. 3, pp. 78-84, 2013.
[12]O. A. Wahab, H. Otrok, and A. Mourad, "A cooperative
watchdog model based on Dempster–Shafer for detecting
misbehaving vehicles," Computer Communications, vol.
41, pp. 43-54, 2014.
0
10
20
30
40
50
60
70
80
1 2 3 4 5
Accuracy(%)
Node Event Detection
Proposed System
Existing System

[13]J. Sengathir, R. Manoharan, and R. R. Kumar,
"Markovian process based reputation mechanisms for
detecting selfish nodes in MANETs: A survey," in Fifth
International Conference on Advanced Computing
(ICoAC), 2013 2013, pp. 217-222.
[14]M. Nikmaram, "An energy efficient acknowledgement-
based method for selfish node detection and avoidance in
open MANET," Universiti Teknologi Malaysia, Faculty
of Computing, 2014.
[15]A. K. Akhtar and G. Sahoo, "Classification of selfish and
regular nodes based on reputation values in MANET
using adaptive decision boundary," 2013.

Detecting Misbehaving and Selfish Nodes in the Network using Watchdog Mechanism

More Related Content

What's hot (19)

Similar to Detecting Misbehaving and Selfish Nodes in the Network using Watchdog Mechanism (20)

Recently uploaded (20)

Detecting Misbehaving and Selfish Nodes in the Network using Watchdog Mechanism