International Journal of Advanced Research in ISSN : 2347 - 8446 (Online)

Computer Science & Technology (IJARCST 2016) Vol. 4, Issue 2 (Apr. - Jun. 2016) ISSN : 2347 - 9817 (Print)

Discover and Prevent the Sinkhole Attacks in Wireless

Sensor Network using Clustering Protocol
I
Vishwas D B, IIChinnaswamy C.N, [Link]
PG Student, CNE, II Associate Professor, III Professor
I

I,II,III
Dept. of IS&E, The National Institute of Engineering, Mysore, India

Abstract
Wireless sensor network (WSN) is one of the new network techniques and it consists of any small wireless sensor nodes. These nodes
are capable of sensing and send data to base stations. Sinkhole attacks are work by making a compromised node look especially
attracts network traffic by advertising fake routing updates. This fake updates are creates opportunities for attackers. So we want to
detect and avoid sinkhole attack, many different methods are used to detect and avoid sinkhole attacks.
I propose a method to detect and avoid the sinkhole attacks for clustering technology in wireless sensor networks. Initially network
is divided into number of clusters, every cluster has one header and the header is directly communicated to destination. In HEED
clustering protocol the header nodes are not selected randomly. We examine our work in terms of throughput, packet loss rate, and
end to end delay.

Keywords
Sinkhole, Clustering protocol, HEED, Aggregation, Mobile Agent.

I. Introduction for communicating with their own members and other cluster
Wireless sensor network (WSN) is a system of network spatially heads. There are diverse methods for selecting cluster head. In
distributed devices using wireless sensor nodes to sense some methods cluster head is selected by cluster members; while,
environmental or physical conditions. in other methods cluster head is selected by network designer.
The Individual nodes are competent of sensing their environments Cluster head may either remain constant during network lifetime
and sending data to one or more compilation or change based on the algorithm. The same is true for members
Points in a WSN. One of the most significant issues for WSNs is of clusters. In clustering methods each sensor is either a cluster
efficient data transmission. head which introduces itself in a specific region or is a member
which must introduce itself to cluster head and become its member.
A. Sinkhole attack The members are only able to communicate with their own cluster
Sinkhole attacks typically work by making a compromised node head and transmit.
look especially attractive to surrounding nodes with respect to the
fake routing algorithm. Sinkhole attack attempts to harm WSN’s
where in an intruder attempts to redirect all network traffic towards
itself, by providing falls routing data. These effects on the network
load balance as well as it provide sausage for other attacks.

Fig 2. Clustered wireless network.

II. Existing System

In existing system the mobile agent technology based security
Fig [Link] of sinkhole attack in wireless sensor network. solution against sinkhole attack for wireless sensor networks
(WSNs). WSN has a dynamic topology, intermittent connectivity,
B. Clustering technology and resource constrained device nodes.
Clustering divides nodes into groups which are called clusters. Mobile agent is a program segment which is self-controlling. They
Each cluster is managed by a cluster head and all members are navigate from node to node not only transmitting data but also
coordinated with the cluster head. Cluster heads are responsible doing computation. They are effective paradigm for distributed

© 2013, IJARCST All Rights Reserved 26 [Link]

ISSN : 2347 - 8446 (Online) International Journal of Advanced Research in
ISSN : 2347 - 9817 (Print) Vol. 4, Issue 2 (Apr. - Jun. 2016) Computer Science & Technology (IJARCST 2016)

applications, and especially attractive in a dynamic network When monitoring cluster-heads, member nodes of a cluster-
environment. This mechanism does not require more energy head take turns to monitor this cluster-head. This mechanism
Sinkhole attacks are work by making a compromised node look reduces the monitor time, and therefore saves the energy of the
especially attracts network traffic by advertising fake routing member nodes. When monitoring member nodes, cluster-heads
updates. This fake updates are creates opportunities for attackers. have the authority to detect and revoke the malicious member
So we want to detect and avoid sinkhole attack, many different nodes clustering is an effective and convenient way to enhance
methods are used to detect and avoid sinkhole attacks. performance of the WSNs system.
For clustering we used HEED protocol this protocol is a
1. Existing Approaches hierarchical, distributed, clustering scheme in which a single-hop
Different peoples are proposed different methods to detect and communication pattern is retained within each cluster, where as
avoid sinkhole attacks .here we discusses these methods. multi-hop communication is allowed among CHs and the BS. Only
We consider some previous researchers papers they may be sensors that have a high residual energy are expected to become CH
classified into rule based, anomaly based, statistical methods, nodes and broadcast a unique value to their authentication keys,
hybrid systems, cryptographic key [Link] to the selected set of nodes in first round of data aggregation. Any
Rule based: In this approach rules are designed based on the node needs to transfer the data within the group, it transfers slices
technique or behaviour used to launch attacks (sinkhole attacks). of data to another nodes in that group, encrypted by individual
These rules are running on each sensor node. Any node will be authentication keys and in receiving end decrypts then combine
considered an isolated and adversary from the sensor network if the slices finally transfers the secure data to its cluster header .The
it attacks the rules. cluster header aggregates and encrypts the data with the shared
Anomaly based: In this method detection and avoid using search secret key of the destination and forwards it to the destination.
the anomalous in the network. Subset of anomaly based detection using clustering technique along with replicated mobile agents
approaches are Rule based and statistical approaches. to prevent the wireless sensor network from sinkhole attacks.
Statistical: In this approaches we recorded data associated and use mobile agents to aware every node from its trusted
with some activities of the sensor nodes in network. Then the neighbouring nodes so they do not listen to the traffics generated
compromised node is detected by using comparing the correct by malicious nodes. We are using clustering technique along
behaviour of the threshold value the values are used as reference, with replicated mobile agents to prevent the wireless sensor
any seniors node are above that threshold value is considered that network from sinkhole attacks and use mobile agents to aware
nodes are an intruders . every node from its trusted neighbouring nodes so they
Cryptographic: In cryptographic approach the dates are protected do not listen to the traffics generated by malicious nodes. We
by using encryption and decryption keys for integrity and examine our work in terms of throughput, packet loss rate, and
authenticity of packets travelling in the whole network. If packets end to end delay.
are transmitted in consider network is encrypted the information Slice method: - When a node wants to send data to its neighbouring
such that to access the information we requires a key and we can nodes, it slices the data into number of pieces (since network size
alter the information can be easily detected. is depends on pieces). It holds the one of the slices with it. The
Hybrid: In hybrid method we combine both the combination of remaining slices are encrypted with their respective authentication
both cryptographic approaches and anomaly approach. Benefits keys and sent to rest of the nodes.
of this approach are can able to catch malicious nodes when their
signature is not included in detection database and we reduced the IV. Literature Survey
false positive rate by using combination of both approaches. Md. Ibrahim Abdullah et. al. [1] presented a technique to
successfully detect the sinkhole attack when this malicious node
III. Proposed System located at far distance from base station. The technique is also
In proposed system I focus on clustering and sinkhole attacks. applicable to wormhole attack as the attack is almost similar
Here instead of mobile agent broadcasting technique. to sinkhole attack. Proposed technique is also applicable when
We use HEED Cluster concept and broadcasting of messages sinkhole nodes advertise high quality link, strong transmitted
done through the cluster heads. power etc. Here we have to sort the advertising parameter and
Clustering is an effective and convenient way to enhance take decision which value is strange.
performance of the WSNs system. Efficient data transmission is In (P. Samundiswary and Dananjayan, 2010)[3] a secured
one of the most important issues for WSNs. Initially network is path redundancy algorithm has been applied to implement in
separated into number of clusters, each cluster is headed by a cluster heterogeneous sensor networks by using alternate path scheme
header and the header is directly connected to destination. The in these networks with mobile nodes for mobile sinks to defend
header upon identifying the detecting nodes selects a set of nodes against sinkhole attacks. This proposed approach is not suitable
randomly and broadcast a unique value to their authentication keys, for homogenous sensor networks.
to the selected set of nodes in first round of data aggregation. In (D Sheela[2] and Mahadevan, 2011) they introduced mobile
When any node within the group needs to transfer the data, it agents to detect the sinkhole attack in wireless sensor networks.
transfers slices of data to other nodes in that group, encrypted by But in this as the number of nodes increases the overhead of the
individual authentication keys. Each receiving node decrypts, sums network also decreases that degrade performance of the network.
up the slices and transfers the encrypted data to the cluster header. In (Sina Hamedheidari and Reza Rafeh, 2013) they provide a
The header aggregates and encrypts the data with the shared secret mobile agent based approach to detect and prevent the sinkhole
key of the destination and forwards it to the destination. The set attacks in wireless sensor networks. They used mobile agents to
of nodes is reselected with new set of authentication keys in the detect malicious nodes and trusted neighbours in order to inform
second round of aggregation. nodes from their environment.

[Link] 27 © All Rights Reserved, IJARCST 2013

International Journal of Advanced Research in ISSN : 2347 - 8446 (Online)
Computer Science & Technology (IJARCST 2016) Vol. 4, Issue 2 (Apr. - Jun. 2016) ISSN : 2347 - 9817 (Print)

Bhoopathy, V[4] In this paper they propose a securing node References

capture attacks for hierarchical data aggregation in wireless sensor [1] Md. Ibrahim Abdullah “Detecting Sinkhole Attacks in Wireless
networks. Initially network is separated into number of clusters, Sensor Network using Hop Count” Computer Science and
each cluster is headed by an aggregator and the aggregators are Engineering, Islamic University, Kushtia, Bangladesh
directly connected to sink. When any node within the group needs [2] [Link] Naveen kumar “A NON CRYPTOGRAPHIC
to transfer the data, it transfers slices of data to other nodes in that METHOD OF SINK HOLE ATTACK DETECTION IN
group, encrypted by individual authentication keys. Each receiving WIRELESS SENSOR NETWORKS” . Computer Science
node decrypts, sums up the slices and transfers the encrypted data and Engg Computer Science India Bangalore, India
to the aggregator By simulation results, we demonstrate that the [3] Daniel Dallas, Christopher Leckie, Kotagiri Ramamohanarao
proposed technique resolves the security threat of node capture “Hop-Count Monitoring: Detecting Sinkhole Attacks in
attacks. Wireless Sensor Networks” NICTA Victoria Research
Haowen Chan et al [5] Secure hierarchical in-network data Laboratory Department of Computer Science and Software
aggregation is guaranteed to identify any manipulation of the Engineering The University of Melbourne, Australia
aggregate by the adversary beyond what is achievable through [4] Bhoopathy, V. “Securing Node Capture Attacks for Hierarchical
direct injection of data values at compromised nodes. In other Data Aggregation in Wireless Sensor Networks” International
words, the adversary can never gain any advantage from Journal of Engineering Research and Applications (IJERA).
misrepresenting intermediate aggregation computations. The Department of CSE, Annai Mathammal Sheela Engineering
system incurs only O(Δlog2 n) node congestion, supports arbitrary College, Tamil Nadu, India
tree-based aggregator topologies and retains its resistance against [5] Haowen Chan et al “Secure Hierarchical In-Network
aggregation manipulation in the presence of arbitrary numbers Aggregation in Sensor Networks” Carnegie Mellon
of malicious nodes. The main algorithm is based on performing University
the SUM aggregation securely by first forcing the adversary to [6] Radhikabaskar, [Link] Raja, et al “Sinkhole
commit to its choice of intermediate aggregation results. Attack Detection In Hierarchical Sensor Networks” .IEEE-
Sanjeev Kumar Gupta[6] In WSNs there is one mechanism used International Conference on Recent Trends in Information
to enlarge the lifespan of network and provide more efficient Technology, ICRTIT 2011 978-1-4577-0590-8/11/$26.00
functioning procedures that is clustering. Clustering is a process ©2011 IEEE MIT, Anna University, Chennai. June 3-5,
to subdivide the sensing field of sensor network into number of 2011
clusters. Each cluster selects a leader called cluster head. A cluster
head may be elected by the sensor node in the cluster or pre
assigned by the network designer. Optimized Clustering can save
lot of energy in the network. In our paper we have surveyed various
clustering protocols for wireless sensor networks and compared on
various parameters like cluster count, cluster size, cluster density,
message count, node deployment, heterogeneity of nodes, location
awareness and cluster head selection process etc.
Radhikabaska et [6] proposed an intrusion detection system against
sinkhole attack in wireless sensor networks with mobile sink. This
scheme is based on a hierarchical topology to secure any cluster-
based routing protocols. signature technique that represents the
detection data rate of a we can hinder deploying a replicated false
mobile sink .Cell leaders activate their IDS only when sinkhole
event occurs. This permits to reduce the number of nodes running
their IDS and minimize energy consumption.

V. Conclusion
The main aim is avoiding sinkhole attack in WSN, sinkhole attack
means duplicate nodes confuse and attract the nodes of entire
network and reduce the network performance .Therefore we have
Securing Node Capture Attacks for hierarchical data aggregation
.The aggregator aggregates and encrypts the data with the shared
secret key of the sink and forwards it to the sink.
Saving the energy of the nodes and reduce time consuming of
network by using HEED clustering protocol. In HEED clustering
protocol the header nodes are not selected randomly and using
clustering technique along with replicated mobile agents to
prevent the wireless sensor network from sinkhole attacks. Mobile
agents are software entities which can function autonomously in a
particular environment. They are able to carry out some activities
in a flexible and intelligent manner.

© 2013, IJARCST All Rights Reserved 28 [Link]