A THRESHOLD BASED AUTHENTICATION MECHANISM FOR VEHICULAR AD HOC NETWORK (VANET)

https://iaeme.com/Home/journal/IJEET 376 editor@iaeme.com
International Journal of Electrical Engineering and Technology (IJEET)
Volume 11, Issue 9, November 2020, pp. 376-387, Article ID: IJEET_11_09_035
Available online at https://iaeme.com/Home/issue/IJEET?Volume=11&Issue=9
ISSN Print: 0976-6545 and ISSN Online: 0976-6553
DOI: https://doi.org/10.17605/OSF.IO/AU7XP
© IAEME Publication Scopus Indexed
A THRESHOLD BASED AUTHENTICATION
MECHANISM FOR VEHICULAR AD HOC
NETWORK (VANET)
Dr. B. Karthikeyan
Assistant Professor, Department of Computer Science, Bishop Heber College (Autonomous)
(Affiliated to Bharathidasan University, Tiruchirappalli), Tiruchirappalli, Tamilnadu, India.
ABSTRACT
Since Vehicular ad hoc networks (VANETs) are vulnerable to various kinds of
attacks, there is a need to fulfill the security requirements like message privacy,
integrity, and authentication. The authentication technique is said to be efficient if it
detects compromised nodes accurately with less complexity, reduced authentication
delay, and keying overhead. In this paper, a threshold-based authentication scheme for
cluster-based VANETs is proposed. The vehicles are clustered, and the threshold degree
of each node is estimated. The threshold degree is a combination of direct threshold
degree and indirect threshold degree. Based on this estimated threshold degree, cluster
heads are selected. Then, each vehicle is monitored by a set of verifiers, and the
messages are digitally signed by the sender and encrypted using a public/ private key
as distributed by a threshold authority and decrypted by the destination. This verifies
the identity of sender as well as receiver thus providing authentication to the scheme.
By simulation results, we prove that the proposed technique provides high security with
less overhead and delay.
Key Words: Vehicular Ad Hoc Network, Clustering, Threshold Authentication,
Threshold Authority, Monitoring
Cite this Article: B. Karthikeyan, A Threshold Based Authentication Mechanism for
Vehicular Ad HOC Network (VANET), International Journal of Electrical Engineering
and Technology (IJEET), 11(9), 2020, pp. 376-387.
https://iaeme.com/Home/issue/IJEET?Volume=11&Issue=9
1. INTRODUCTION
VANET is made up of extremely mobile automobiles with sparingly installed stations at the
sides of the road; all of them provided with gadgets as well as sensing devices in some cases,
that communicate wirelessly. By making use of vehicle-to-vehicle (V2V) ad hoc mode as well
as between vehicles and roadside stations by means of vehicle- to-road (V2R) or vehicle-to
infrastructure (V2I) communication mode through a base station (BS) or access point (AP),
wireless communication can be achieved. For this communication to take place, the AP is
usually deployed down the road contained by the BS or AP range for transmission [1]. On board

A Threshold Based Authentication Mechanism for Vehicular Ad HOC Network (VANET)
units (OBUs) are deployed on these automobiles in order to enable them and the units along the
road, comprising the infrastructure connecting the vehicular network to the central unit.
VANET facilitates data transmission such as messages indicating caution related to road
situation, traffic condition, and driving condition of the drivers. Application of VANET include
accumulating, processing, allocating and delivering the information about the road in real time
[2–5].
The increased movement of the vehicles as a result of the repeatedly altering topology
imposes a crucial task in delivering unicast communication among vehicles itself or between
vehicles and the concerned infrastructure [4, 6].
With the increase in distance, the energy required to provide good quality communication
also increases. As a result, the overall energy consumed by the transceiver will be high. On the
basis of the number of the relaying nodes and transmission distance between every pair of
nodes, the energy consumed increases during communication in multi-hop VANET. Therefore,
the energy required for a single transmission amplifies nonlinearly, in the case of little hops and
higher transmission distances. So, to obtain the best energy efficiency, we need to maintain a
tradeoff between the hop number and the transmission range for every hop [3, 7].
The target of VANET is achieving higher level of safety on the road. In order to achieve it,
every vehicle working as a sensor sends information to each other like warnings related to the
present speed, physical location and ESP activity, which lets the drivers to take appropriate
measures in case of hazardous condition like accidents, traffic problem, and glaze. Also, official
vehicles used by the police and the firefighters can make use of it to transfer messages for
stopping other vehicles or clearing the road. Moreover, services on the basis of location and
Internet along the road can be provided by VANET. With regard to the protection concerns,
reliability, privacy, and accessibility that are the safety and confidentiality requirement, it is
required for the three application divisions like warnings and telematics information, alarm
signals and instructions, and value-added services. There is a need of a secure topology
maintaining threshold and allowing cryptography process [8, 9]. Jamming, impersonation,
privacy violation, forgery, in-transit traffic tampering, onboard tampering, and so forth are the
situations to which VANETs are vulnerable. Therefore, there is a need for VANET to fulfill the
security requirements like message privacy and integrity, message non negation, unit
validation, admission management, secrecy, accessibility, and responsibility identification [10,
11].
2. PROBLEM IDENTIFICATION
Achieving energy proficiency as well as security is a challenge in VANET. With the help of
cryptographic theory [4], signature using cryptography [12], privacy preservation [13],
threshold models [14, 15], anonymous credential [16], and collaborative protocol [17], the
works [4, 12–17] have guaranteed secure networks. But, certain issues still exist in the current
network such as power consumption [3], incapacity to discover compromised nodes [4],
complexity [12], message dropping [13], higher delay [14], overhead [15], and collision [17].
Hence, an objective is to develop a scheme in VANET with ability to detect compromised
nodes, less complexity, reduced message dropping, delay, overhead, and collision.
By using the clustering technique and the key distribution mechanism, security can be
accomplished in VANET, where the vehicles are gathered together in clusters and the
problematic vehicles are secluded by a particular algorithm [12]. Later, on the basis of the proxy
signature which is encrypted and transmitted through a safe channel, keys are produced. But,
this mechanism is very complicated, and there are possibilities for the VANET to break down,
on high rate of network utilization leading to reduced energy. The privacy and integrity
requirement has not yet been fulfilled in VANET.

B. Karthikeyan
3. THRESHOLD BASED AUTHENTICATION TECHNIQUE
Aim of this paper is to develop a threshold based authentication scheme for cluster based
VANETs. In this scheme the vehicles are clustered [12] and the threshold degree of each node
is estimated. The threshold degree is a combination of direct threshold degree and indirect
threshold degree. Direct threshold degree of node is calculated from neighbors using past
interactions whereas indirect threshold degree is recommendation threshold degree from the
most similar nearest neighbors. Based on this estimated threshold degree, the cluster heads (CH)
are selected. Then each vehicle is monitored by a set of verifiers. Then we add digital signature
to the messages signed by the sender and encrypted using a public/ private key as distributed
by a threshold authority and decrypted by the destination. This verifies the identity of sender as
well as receiver thus providing authentication to the scheme.
Figure 1 Block Diagram for Threshold based Authentication scheme
3.1. Adversarial Model
The attacks in VANET are of two types. They are active attack and passive attack. In a passive
attack, the attacker eavesdrops but does not modify the message, whereas in an active attack,
the attacker may transmit messages, replay old messages, modify messages in transit, or delete
selected messages. Man-in-the-middle and replay attacks are considered in the proposed work.
Man-in-the-middle attack is an active attack in which the attacker secretly relays and alters the
communication between two parties who believe that they are directly communicating with
each other. Replay attack is also an active attack in which the attacker may repeat the data or
delay the data. Node-to-node authentication (described below) is used to address these attacks
(Fig. 2).
3.2. Clustering of Vehicle
We assume that there are several Certification Authorities (CAs) in the network, where each
CA can authenticate all the vehicles located inside its region. A CA is a threshold third party
that manages identities, cryptography keys, and credentials of vehicles.
Initially the vehicles are divided into several clusters in a highway environment with two
bands and each band having three lanes. Each cluster consists of one cluster head (CH) and one
or more members. Vehicles in one cluster are linked directly and vehicles that are located in
two different clusters can communicate together via their CHs. Each vehicle can play the role
of a CH or gateway or member. If one vehicle is located within two or more clusters, it is called

a gateway. Each CH maintains the information about its members and gateways. The cluster
head election process is described in Algorithm 1.
Figure 2 Cluster Formation and Cluster Head Selection
Table 1 Algorithm notation and its description
Notation Description
Vi Each Vehicle in the network, i=1,2,3……….
Vj Neighbor of Vi
Addi Address of Vi
Id Id of Vi
Nlj Neighbor list of Vj
Dij Distance between Vj and Vi
NVj Number of neighbors of Vj
R Dynamic transmission range
𝜃 Direction of vehicle
S Speed of the vehicle
DTr Threshold degree
𝛼, 𝛽, 𝛿, 𝛾, 𝜂 Weighting Constants
Step 1: Each vehicle Vi declares itself as a CH and broadcast the bacon B[addi, Idi]
Step 2: Each vehicle Vj creates Nlj after receiving B[addi, Idi] from each Vi.
Step 3: Then Vj estimates Dij.
Step 4: Vj calculates a weighted sum
𝑊
𝑗 = 𝛼. 𝑁𝑉
𝑗 + 𝛽. 𝑅 + 𝛿. 𝜃 + 𝛾. 𝑆 − 𝜂. 𝐷𝑇𝑟 (1)
Step 5: The parameters used in the Eq. (1) are calculated by the vehicle. The weighted constants
range from 0 to 1. As the weighted sum is calculated based on these parameters, the CH which
is selected based on it will be threshold and efficient.
Step 5: Then Vk with Wk = Minimum is selected as CH.
3.3. Estimation of Threshold Degree
Threshold degree estimation is done for the selection of Cluster Heads (CH). Threshold
relationships made from the direct interactions is described as direct threshold. The threshold

B. Karthikeyan
relationship built from the trust node or the chain of trust node is called as indirect threshold
node [14]. The direct threshold degree from vehicle p to vehicle q is given by,
𝑇𝑛𝑒𝑤
𝑜𝑙𝑑 (𝑝, 𝑞) = {
𝑇𝑛𝑒𝑤
𝑜𝑙𝑑 (𝑝, 𝑞) + 𝑅𝐹, (𝑆𝑇 > 0)
𝑇𝑛𝑒𝑤
𝑜𝑙𝑑 (𝑝, 𝑞) − 𝑅𝐹, (𝑆𝑇 > 0)
(2)
where
Told = Previous threshold degree (i.e., the value calculated during previous CH selection
process).
RF—Reward factor,
PF—Penalty factor,
ST, FT—Number of successful and failed transactions between Told and Tnew in time interval
Δ𝑡.
The indirect threshold degree from vehicle p to vehicle q is given by,
K—common neighbor vehicle
s(p,k)—similarity of values of vehicle p and k.
m—number of most similar nearest-neighbors of p and q.
The estimation of threshold degree is the sum of direct threshold and indirect threshold,
𝛼 𝑎𝑛𝑑𝛽 – weighting factors for 𝑇𝑑 (𝑝, 𝑞) 𝑎𝑛𝑑 𝑇𝑟 (𝑝, 𝑞)
Table 2 Notation and Description of the algorithm2
T(p,q) Threshold degree between vehicle p and q
N(p) Neighbor of node p
𝑇𝑑 Direct threshold degree
𝑇𝑟 Indirect threshold degree
tc Current time
Step 1: Node p collects the local topology information.
Step 2: Td
is calculated by p based on the neighbor table and historical events with N(p) using
(2).
Step 3: If there is no interaction between the p and q, then
Step 4: T(p,q) = Td
Step 5: Store Td
and tc in local information table
Step 6: End if
Step 7: If there is interaction between p and q, then
Step 8: Update Td
Step 9: Tr
is calculated by p by estimating similar Td
values of N(p) to q, using (3)
Step 10: Calculate T(p, q) using (4)
Step 11: End if

3.4. Vehicle Monitoring
In monitoring phase, a set of verifier nodes collect information about the behavior of all vehicles
in a cluster. A vehicle Vi can be a verifier of another vehicle Vj if T(Vi)>T(Vj), where T is the
total threshold degree stored in the neighbor table of each node. Let Tmin be the minimum
threshold value of threshold degree. The steps involved in the vehicle monitoring process are
illustrated in Algorithm 3.
Table 3 Notation and its description of algorithm3
Tmin Minimum threshold value of threshold degree
T(Vj) Total threshold degree of vehicle Vj
CA Certificate authority
RSU Road side unit
Figure 3 Vehicle Monitoring
Step 1: {Vj} detect the abnormal behaviors of vehicle Vj by monitoring, when Vj acts as a relay
node or source node.
Step 2: After detecting abnormal behavior of Vj, the CH requests for the threshold degree of Vj
from other verifiers in the cluster.
Step 3: When T(Vj) is different from its old value, the new value of T(Vj) is informed by the
CH to the other cluster members.
Step 4: All other cluster members updates their neighbor table based on the new value of T(Vj).
Step 5: new T(Vj)>= Tmin, then All other cluster members cooperate with Vj.
Step 6: Else
Step 7: CH informs the id of Vj to the CA.
Step 8: CA broadcasts the ID of Vj to all the vehicles and RSUs.
Step 9: End if
3.5. Node to Node Authentication
Initially, we assume that public/private key pairs and certificates are distributed to legitimate
nodes who wish to join the ad hoc network. The keys can be entered manually or through secure
transfer protocols. The messages sent by a vehicle can be protected using digital signature (DS).
The sender attaches a DS at the end of every control message. The DS consists of a value that

B. Karthikeyan
is known by the signer and the content of the message being signed. The sender signs the
message using the private key and the receiver verifiers the message with the signer’s public
key [15].
During the authentication procedure, the node attempting to authenticate presents its
identity and certificate to the authenticating node. The authenticating node will first verify the
certificate using the public key of CA and then challenge the initiating node by encrypting a
nonce with the initiating node’s public key, to test whether it has the corresponding private key.
At the end of the handshake, two nodes exchange secret keys (encrypted with other’s public
key) for quick re-association in the future.
Figure 4 Node to node authentication
3.6. Steps involved in Threshold based Authentication
The entire steps involved in the threshold based authentication technique can be summarized
as:
(a) Initially the vehicles are clustered.
(b) Threshold degree of each node is estimated based on direct and indirect threshold degrees.
(c) In each cluster, cluster head is selected based on the weighted sum.
(d) Vehicles are monitored by a set of verifiers in each cluster.
(e) The threshold degrees of vehicles with abnormal behavior are checked by CH.
(f) Abnormal nodes with least threshold degree are isolated by the CA.
(g) In node to node authentication, a digital signature is added to the messages signed by the
sender and encrypted using a public/ private key as distributed by a trust authority and decrypted
by the destination.
(h) The sender signs the message using the private key and the receiver verifiers the message
with the signer’s public key.

(i) At the end of the handshake, two nodes exchange secret keys for quick re-association in the
future.
4. SIMULATION RESULT AND DISCUSSION
The proposed Threshold based Authentication Technique (TBAT) is compared with Secure
scheme based on Clustering and Key Distribution (SCKD) [12] and VSPN [16]. The
performance is evaluated in terms of Packet delivery ration, authentication delay, key overhead
and detection accuracy.
Table 4 Parameter Setting for simulation
Number of nodes 72
Area size 2500 9 700 m
Number of BANDS 2
Number of lanes per band 3
Radio range 250, 300, 350 and 400 m.
Simulation time 50s
Packet size 512 bytes
Antenna Omni Antenna
Figure 5 Attackers versus authentication delay
Figure 6 Attackers versus delivery ratio

B. Karthikeyan
Figure 7 Attackers verus keying overhead
Figure 8 Attackers versus detection accuracy
Table 5 Number of clusters for various ranges
Range Number of clusters per lane Number of nodes per
cluster
250 3 12
300 3 12
350 2 18
400 2 18
Figure 9 Range verus authentication delay

Figure 10 Range verus delivery ratio
Figure 11 Range verus overhead
Figure 12 Range verus Detection accuracy
5. CONCLUSION
In our paper we developed a threshold based authentication scheme for cluster based VANETs.
For that, the vehicles are clustered and the threshold degree of each node is estimated. The
threshold degree is a combination of direct threshold degree and indirect threshold degree.
Based on this estimated threshold degree, the cluster heads (CH) are selected. Then each vehicle
is monitored by a set of verifiers. Then we add digital signature to the messages signed by the
sender and encrypted using a public/private key as distributed by a trust authority and decrypted

B. Karthikeyan
by the destination. This verifies the identity of sender as well as receiver thus providing
authentication to the scheme. Simulation results show that the proposed technique reduces the
authentication delay and keying overhead while increasing the packet delivery ratio.
REFERENCES
[1] Network simulator, http://www.isi.edu/nsnam/ns.
[2] Qin, H., Li, Z. Wang, Y., Lu, X., Zhang, W. S., & Wang, G. (2010). An integrated network of
roadside sensors and vehicles for driving safety: Concept, design and experiments. In IEEE
International Conference on Pervasive Computing and Communications (PerCom).
[3] Feng, W., Alshaer, H., & Elmirghani, J. M. H. (2010). Green information and communication
technology: Energy efficiency in a motorway model. IET Communications, 4(7), 850–860.
[4] Pradeep, B., Manohara Pai, M. M., Boussedjra, M., & Mouzna, J. (2009). Global public key
algorithm for secure location service in VANET. In 9th International Conference on Intelligent
Transport Systems Telecommunications, (ITST).
[5] Rivas, D. A., Barcelo-Ordinas, J. M., Zapata, M. G., & Morillo- Pozo, J. D. (2011). Security on
VANETs: Privacy, misbehaving nodes, false information and secure data aggregation. Journal
of Network and Computer Applications, 34(6), 1942– 1955.
[6] Nayyar, Z., Khattak, M. A. K., Saqib, N. A., & Rafique, N. (2015). Secure clustering in
vehicular ad hoc networks. International Journal of Advanced Computer Science and
Applications (IJACSA), 6(9), 285–291.
[7] Feng, W., & Elmirghani, J. M. H. (2009). Green ICT: Energy efficiency in a motorway model.
In Third International Conference on Next Generation Mobile Applications, Services and
Technologies.
[8] Plo¨ ßl, Klaus, & Federrath, Hannes. (2008). A privacy aware and efficient security
infrastructure for vehicular ad hoc networks. Computer Standards and Interfaces, 30, 390–397.
[9] Mokhtara, Bassem, & Azab, Mohamed. (2015). Survey on security issues in vehicular ad hoc
networks. Alexandria Engineering Journal, 54(4), 1115–1126.
[10] Qian, Y., & Moayeri, N. (2008). Design secure and application oriented VANET. In IEEE
Vehicular Technology Conference, VTC Spring.
[11] Fathian, M., & Jafarian-Moghaddam, A. R. (2015). New clustering algorithms for vehicular ad-
hoc network in a highway communication environment. Wireless Networks, 21(8), 2765–2780.
[12] Daeinabi, A., & Rahbar, A. G. (2013). An advanced security scheme based on clustering and
key distribution in vehicular adhoc networks. Computers and Electrical Engineering.
[13] Ganãń, C., Munõz, J. L., Esparza, O., Mata-Dıáz, J., & Alins, J. (2014). PPREM: Privacy
preserving REvocation mechanism for vehicular ad hoc networks. Computer Standards and
Interfaces, 36, 513–523.
[14] Zhizhong, J., Chuanhe, H., Liya, X., Bo, W., Xi, C., & Xiying, F. (2012). A trusted opportunistic
routing algorithm for VANET. In IEEE Third International Conference on In Networking and
Distributed Computing (ICNDC), pp. 86–90.
[15] Chen, T., Mehani, O., & Boreli, R. (2009). Trusted routing for VANET. In IEEE 9th
International Conference on Intelligent Transport Systems Telecommunications, (ITST).
[16] Chim, T. W., Yiu, S. M., Hui, L. C. K., & Li, V. O. K. (2014). VSPN: VANET-based secure
and privacy-preserving navigation. IEEE Transactions on Computers, 63(2), 1–14.
[17] Barba, C. T., Aguiar, L. U., Igartua, M. A., Parra-Arnau, J., Rebollo-Monedero, D., Forne´, J.,
et al. (2013). A collaborative protocol for anonymous reporting in vehicular ad hoc networks.
Computer Standards and Interfaces, 36, 188–197.

[18] Subhashini, M., & Gopinath, R., Mapreduce Methodology for Elliptical Curve Discrete
Logarithmic Problems – Securing Telecom Networks, International Journal of Electrical
Engineering and Technology, 11(9), 261-273 (2020).
[19] Upendran, V., & Gopinath, R., Feature Selection based on Multicriteria Decision Making for
Intrusion Detection System, International Journal of Electrical Engineering and Technology,
11(5), 217-226 (2020).
[20] Upendran, V., & Gopinath, R., Optimization based Classification Technique for Intrusion
Detection System, International Journal of Advanced Research in Engineering and Technology,
11(9), 1255-1262 (2020).
[21] Rethinavalli, S., & Gopinath, R., Botnet Attack Detection in Internet of Things using
Optimization Techniques, International Journal of Electrical Engineering and Technology,
11(10), 412-420 (2020).
[22] Priyadharshini, D., Poornappriya, T.S., & Gopinath, R., A fuzzy MCDM approach for
measuring the business impact of employee selection, International Journal of Management
(IJM), 11(7), 1769-1775 (2020).
[23] Poornappriya, T.S., Gopinath, R., Application of Machine Learning Techniques for Improving
Learning Disabilities, International Journal of Electrical Engineering and Technology (IJEET),
11(10), 392-402 (2020).

A THRESHOLD BASED AUTHENTICATION MECHANISM FOR VEHICULAR AD HOC NETWORK (VANET)

More Related Content

Similar to A THRESHOLD BASED AUTHENTICATION MECHANISM FOR VEHICULAR AD HOC NETWORK (VANET) (20)

More from IAEME Publication (20)

Recently uploaded (20)

A THRESHOLD BASED AUTHENTICATION MECHANISM FOR VEHICULAR AD HOC NETWORK (VANET)