Volume 6, Issue 2, February – 2021 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Two-way Authentication Algorithm for

Sinkhole Attack Isolation in WSN
Semagn Shifere Andebet Dessiewu Tizazu Bayih
Department of Computer Science Department of Information Technology Department of Information Technology
Woldia University, Ethiopia Woldia University, Ethiopia Woldia University, Ethiopia

Abstract:- Wireless sensor networks (WSN) are battery with a short lifespan, and because of physical
networks that are self-configuring and run without any constraints, their replacement is impractical.Furthermore,
central coordinator. Security and energy consumption any number of sensor nodes must be able to scale up the
are a big problem for wireless sensor networks because design and protocol of the sensor networks. As the lifetime
of these network properties. The malicious nodes that of the battery can be extended on the off chance that one can
join the network cause security attacks.In terms of find out how to minimize the contact calculation(Dr. G.
active and passive attacks, it is possible to distinguish Padmavathi, Mrs. D. Shanmugapriya, 2009). Consumption
attacks. In the active form of attack, the sinkhole assault. can be minimized in the sensing energy subsystem by using
The malicious nodes in the sinkhole attack spoof the low-power hardware components for energy-efficient
base station's identity and behave like the base station. routing protocols, such as Hybrid Energy-Efficient
Instead of base stations, the sensor nodes start Distributed (HEED).Clustering involves grouping nodes
transmitting data to malicious nodes. Cluster heads are into clusters and regularly selecting cluster heads so that
chosen on the basis of energy, distance, to minimize people from a cluster can talk to their cluster heads, and
energy consumption of the network LEACH protocol, these cluster heads send to a base station aggregated data
which divides the entire network into clusters and obtained from their individuals.There is a cluster head in
clusters. Therefore, the total life of the network is each cluster, and from that cluster there are individual rest
decreased and the amount of energy consumed is nodes. Clustering results in a two-level order in which the
increased when a sinkhole attack occurs.In this research higher level is formed by cluster heads while component
paper, the latest method is designed to identify and nodes frame the lower level.(G.H. Raghunandan, IEEE
isolate malicious nodes from the network. This 2011). Compared to component nodes, because the cluster
algorithm works by using recognition to quickly classify head constantly transmits data over longer separations, they
attacks. The proposed algorithm is implemented in NS2 lose more energy. The clustering technique is used to
and those parameters are evaluated in terms of minimize the consumption of resources. LEACH is the
performance. As opposed to current techniques, our protocol that is the most powerful wireless sensor network
proposed technique performs well in terms of all clustering protocol.The cluster heads in the LEACH
parameters. protocol are selected randomly on the network. The head of
the cluster gets its sensor nodes based on size. Under the
Keywords:- Active, Attack, Sinkhole, Malicious, LEACH, cluster head will come the nodes which are nearest to the
WSN cluster head. On the basis of energy, the clusters are
randomly modified.The cluster heads in the LEACH
I. INTRODUCTION protocol are picked at random on the network. Based on
size, the cluster head gets its sensor nodes. Under the cluster
As of late, cellular network infrastructure and mobile head, the nodes that are nearest to the cluster head will
communications have seen a flourishing evolution. A come. On the basis of energy, the clusters are altered
wireless sensor network is the array of various sensing randomly.Safety assaults are commonly known as active
devices in such a way that information relevant to the and passive assaults. Active attacks are those that
surrounding environment of a particular region can be significantly decrease the efficiency of the network in terms
known. They have the potential to change life and the of different parameters. Passive attacks are those that do not
complexities of system building. Different programs include impact the efficiency of the network, but can cause active
sensor-based location detection with sensor networks, attacks in the future.
personal health screen and movement detection(J.-C. Wang,
C.-H. Lin, E. Siahaan, B.-W. Chen, and H.-L. Chuang, A wormhole attack enables an attacker to form a
2014). Within these networks, the size of multiple tunnel between two distant WSN locations, and the packets
limitations such as battery size, processors, memory storage are transmitted through an in-band or out-of-band channel
of information and so on is significant. With the assistance using that tunnel. (Dong D, 2008). Thus, the wormhole
of different optimization algorithms, the use of energy must tunnel forms a pair of attackers. A misapprehension that
be progressed within the networks.Inside the detected and they are similar to each other makes two distant nodes;
routing data sent through the WSNs, different time however, it is not so in reality. The current wormhole will
constraints are present. Sensor nodes typically depend on a

IJISRT21FEB271 www.ijisrt.com 1450

Volume 6, Issue 2, February – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
then draw a significant amount of network traffic to WSN the efficiency of the proposed algorithm was examined. The
and move through it. findings have shown the algorithm's efficacy and accuracy.
They also say that for wireless sensor networks, their
Blackhole and Grayhole: In this attack, during the overheads for communication and computation are
path-finding phase or in route update messages, a malicious relatively low.
node falsely advertises good paths to the destination node.
The malicious node could have the aim of hindering the According to the device that detects the presence of a
path-finding process or intercepting all the data packets sent sinkhole attack is proposed in conjunction with novel
to the appropriate destination node. The grayhole attack is intrusion detection.(Daniel Dallas, 2016). The scheme is
regarded as a more delicate form of this attack, where the focused on control of the hop count. The ADS is easy to
malicious node drops the data packets intermittently, implement with a small footprint, since the hop-count
making their detection much more difficult. function is easily obtained from routing tables.In addition,
the proposed ADS is applicable to any routing protocol
Sinkhole Attack: An attacker makes a compromised which, as a measure of distance between source and
node appear more appealing to its neighbors in a sinkhole destination nodes, dynamically maintains a hop-count
attack by forging the routing information. The result is that parameter. In a simulated network, the scheme can detect
the neighboring nodes select the compromised node to route attacks with 96 percent accuracy and no false alarms using a
their data via the next-hop node.Selective forwarding is very single detection system.
easy for this form of attack, since all traffic from a wide area
of the network will pass via the compromised node. For According to Su, et al. (C.-C. Su, 2010), two methods
these reasons, the researcher wants to work with a two-way have been suggested to enhance cluster protection for sensor
authentication algorithm to build WSN's energy-efficient networks using IDS. The first technique uses an
and more effective sinkhole isolation algorithm that authentication-based model, which can withstand external
improves network performance. attacks. Adding a message authentication code (MAC) for
each message is the basic technique.Whenever a node
wishes to send a message, a timestamp is applied to it and a
II. RELATED WORK MAC is created by a key pair or independently based on the
sender's key position. The authentication mechanism is used
Research works performed by various researchers and by LEAP so that the recipient can verify the sender. Energy
connected to these works are presented in this section. There saving is called the second scheme.The emphasis of this
are a number of works in WSN that are carried out in the method is the identification of wrongdoing in both Member
field of protection. Nodes (MN) and cluster head nodes (CH). The CH
broadcasts a warning message encrypted with the cluster
According to Annie Mathew et.al, within a sensor key to restrain this particular node when wrongdoing is
network, the accumulation of large numbers of sensor nodes detected.
becomes easy to sense and communicate with in the shortest
range.(Terence, 2017). Security is also a major concern in According to Kavita Tandon (Tandon, August 2016),
the wireless sensor network, due to its communication several routing and security issues in WSNs based primarily
capabilities, among other major problems. There are on Sinkhole attacks were introduced in their paper. It also
different attacks that affect the activity of sensor nodes. In provides different methods for detecting and preventing
WSN, Sinkhole is an attack in which the shortest path sinkhole attacks.It finally ends with the countermeasures
between the sink or destination node is shown by the used against this attack. Anomaly detection can be a better
sinkhole node.Many researchers have so far proposed solution if applied with the algorithm that can minimize
various methodologies for the detection of sinkhole attacks. false alarms, according to most of the research paper.
In this article, the author examined and researched the
sinkhole attack and its classification and methods of using III. PROPOSED ALGORITHM
different parameters to detect sinkhole attacks.
The purpose of the study is to limit the effects of
According to Data consistency and network flow sinkhole attacks on network efficiency. The most vulnerable
information approach, the approach presented in(Edith C. H. attack in WSN is the Sinkhole attack on wireless sensor
Ngai, 2006)involves the base station in the detection networks, which prevents the base station from collecting
process, resulting in a high communication cost for the full and unmodified data from its roots. Instead of the base
protocol. The base station floods the network with a request station, cluster heads relay the data to the malicious
message containing the IDs of the affected nodes. The node.The sinkhole attack is the denial of attack type of
affected nodes reply to the base station with a message operation, which decreases the efficiency of the network in
containing their IDs, ID of the next hop and the associated terms of different parameters. The two-way authentication
cost. From the base station, the obtained data is then used to mechanism is based on a novel algorithm. There is a unique
create a network flow graph to define the sinkhole. The identifier at the base station, which is the complex
algorithm is also robust in dealing with malicious Armstrong number.The base station localizes the location of
cooperative nodes that attempt to conceal the actual the node and assigns each node in the network a unique
intruder.Through both numerical analysis and simulations, number. The cluster head will request their identification

IJISRT21FEB271 www.ijisrt.com 1451

Volume 6, Issue 2, February – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
before transmitting the data to the base station.A malicious  Assign unique number ()
node would not be able to present the base station's  The base station generate unique number for each node
identification number to the head of the cluster. To isolate in the network
malicious nodes in the network, the cluster head will  The generate number is the unique Armstrong number
implement multi-path routing. which is complex in nature and difficult to break
 The base station will also send its unique number of
Proposed Algorithm each node in the network
Input: Use finite node numbers to deploy WSN
 Two-way Authentication ()
Output: Malicious Node Detection
 The cluster head ask unique identification number of
base station
 Deploy WSN with the finite number of sensor nodes  If (Base station fails to present unique number)
 Divide the network into clusters of fixed size and pick  Destination node detected as malicious node
the cluster head in each cluster by applying the LEACH
 Else
protocol based on distance, energy
 Authentication complete
 Apply node localization ()  Data transmission starts in the network
 Base station send ICMP message to each node in the
network IV. RESULTS AND DISCUSSION
 The nodes will reply back the hello message on the
basis of received message, base station judge location We implemented the proposed work in NS2 and
of the sensor node analyzed the findings in relation to certain output
parameters by contrasting the proposed work with existing
techniques.

Parameter Values
Type of antenna Omi-directional
Area 800X800meters
No. of nodes 38
Routing Protocol LEACH
Channel type Wireless channel
Packet size 512byte
Mobility model Two ray ground propagation model
Simulation Time 50s
Traffic Type CBR(UDP)
Table 1:- Simulation Parameters

Fig 1:- Comparison of energy consumption

IJISRT21FEB271 www.ijisrt.com 1452

Volume 6, Issue 2, February – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
The energy consumption of the proposed algorithm has been compared to the current algorithm, as shown in figure 1. It has
been studied that the proposed algorithm's energy consumption is lower because of sink hole attack isolation in the network.

Time Existing Algorithm Proposed Algorithm

4 second 21 joules 16 joules

8 second 26 joules 19 joules

14 second 38 joules 24 joules
Table 2:- Energy consumption comparison

Fig 2:- Throughput Comparisons

The performance of the proposed and current throughput. The study can be expanded in the future to
algorithm is compared, as shown in Figure 2, so that it is enhance the detection of sinkhole attacks using the key
analyzed that throughput will be increased at a steady rate exchange mechanism.
due to the isolation of the sink hole attack in the network.
REFERENCES
Time Existing Algorithm Proposed Algorithm
[1]. C.-C. Su, K.-M. C.-H.-F. (2010). The new intrusion
4 second 68 packets 76 packets prevention and detection approaches for clustering-
8 second 140 packets 172 packets based sensor networks . IEEE Wireless
Communications and Networking Conference.
14 second 260 packets 420 packets [2]. Daniel Dallas, C. L. (2016). Hop-Count Monitoring:
Detecting Sinkhole Attacks in Wireless Sensor
Table 3:- Throughput Comparisons Networks. 15th IEEE International Conference on
Networks, ICON, (pp. 176-181).
V. CONCLUSION [3]. Dong D, L. M. (2008). Topological detection on
wormholes in wireless ad hoc and sensor networks.
WSN is the network in this study in which different IEEE Transactions on Mobile Computing, pp. 698–711.
sensor nodes are installed so that it is possible to monitor [4]. Dr. G. Padmavathi, Mrs. D. Shanmugapriya. (2009). A
the surrounding environmental conditions. The sinkhole Survey of Attacks, Security Mechanisms and
attack decreases the efficiency of the LEACH protocol and Challenges in Wireless Sensor Networks. International
the sinkhole attack is detected and isolated by our Journal of Computer Science and Information Security,
approach.We get the output analyzed in terms of energy 4, 1-9.
consumption is minimized by 26 percent and throughput is [5]. Edith C. H. Ngai, J. L. (2006). On the Intruder
increased by 20 percent by conducting experiments in WSN Detection for Sinkhole Attack in Wireless Sensor
including sinkholes in the code.The findings show that the Networks. IEEE International Conference on
proposed malicious node isolation work improves network Communications, 8, p. 33833389.
performance in terms of energy consumption and

IJISRT21FEB271 www.ijisrt.com 1453

Volume 6, Issue 2, February – 2021 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[6]. G.H. Raghunandan, B. L. (IEEE 2011). A Comparative
Analysis of Routing Techniques for Wireless Sensor
Networks. Proceedings of the National Conference on
Innovations in Emerging Technology.
[7]. J.-C. Wang, C.-H. Lin, E. Siahaan, B.-W. Chen, and H.-
L. Chuang. (2014, Feb). Mixed sound event verification
on wireless sensor network for home automation. IEEE
Trans. Industrial Informatics, 10, pp. 803- 812.
[8]. Tandon, K. (August 2016). Sinkhole Attacks in
Wireless Sensor Network Routing: A Survey. Research
Journal of Computer and Information Technology
Sciences, IEEE , 4(8), 4-7.
[9]. Terence, A. M. (2017, April). A Survey on Various
Detection Techniques of Sinkhole Attacks in WSN.
International Conference on Communication and
Signal Processing, 6-8.

IJISRT21FEB271 www.ijisrt.com 1454