VANET Clustering

Temurnikar et al., International Journal on Emerging Technologies 11(4): 173-177(2020) 173
International Journal on Emerging Technologies 11(4): 173-177(2020)
ISSN No. (Print): 0975-8364
ISSN No. (Online): 2249-3255
Development of Multi-Hop Clustering Approach for Vehicular Ad-hoc Network
Ankit Temurnikar
1
, Pushpneel Verma
2
and Jaytrilok Choudhary
3
1
Research Scholar Computer Science & Engineering, Bhagwant University, Ajmer (Rajasthan) India.
2
Assistant Professor, Department of Computer Science & Engineering, Bhagwant University, Ajmer (Rajasthan) India.
3
Associate Professor, Department of Computer Science & Engineering, MANIT Bhopal (Madhya Pradesh), India.
(Corresponding author: Ankit Temurnikar)
(Received 16 May 2020, Revised 22 June 2020, Accepted 03 July 2020)
(Published by Research Trend, Website: www.researchtrend.net)
ABSTRACT: A Vehicular ad-hoc network is a type of mobile ad-hoc network, it’s generally from with the
group of vehicle nodes without using a any predefined infrastructure or network. All the vehicles are
communicated with each other in a wireless transmission; the objective is to provide real time and reliable
communication between the all participating vehicles nodes in a network. In this paper we use the multi-hop
clustering techniques using with the cluster head selection process for the vehicular ad-hoc network, the
cluster head node is selected on the basis of closest speed of node with average speed of node and having a
more network life. Here the challenges is to update the cluster head value and selection of nodes is
dynamically, All the simulation process done with the network simulator compare with the existing
technique, and our proposed method gives better result than the previous techniques, all the results are
obtained with some parameters like network overhead, throughput, end to end delay, packet loss rate and
packet delivery ratio is discussed in this paper we have enhance the previous work by adding two new
parameter percentage of stability and path length with the help QoS matric calculation. The key idea in this
paper to proposes an MHC Approach for VANET that considers Quality of Service (QoS) requirements,
mobility constraints and the distrust value parameters.
Keywords: Vehicular ad-hoc network, Flat Routing, Optimization, Energy efficiency, life-time, QoS, MHC approach.
Abbreviations: MANET Mobile ad-hoc network, VANET Vehicular ad-hoc network, V2VVehicle to vehicle, V2I
Vehicle to infrastructure, DSRC Dedicated short rang communication.
I. INTRODUCTION
In recent years a wireless communication have much
more attention for the researchers, the vehicular ad-hoc
network is one of them because it’s high applicability for
the commercial usage. The vehicular ad-hoc network is
simulated with the network simulator and using some
software for high speed mobile environment such as the
wireless access in vehicular environment (WAVE) and
the Dedicated Short Range Communication (DSRC)
protocol.
Vehicular ad-hoc networks are divided into two forms
V2V (vehicular to vehicular) In which vehicle can
communicate directly with another vehicle and another
types is V2I (vehicle to infrastructure) here vehicles
communicate with using some infrastructure unit such
as Road Side Unit (RSU) as shown in Fig. 1.
Fig. 1. Illustrative network architecture of a vehicular ad-
hoc network.
This can Support many applications such as the road
safety, traffic management and infotainment. The data
link layer parts i.e. medium access control layer is
common medium as responsible to channel utilization in
an efficient manner.
The increased movement of participating vehicle nodes
in a network as a result of repeatedly imposing the
frequently change the topology is very critical task in a
wireless communication among the vehicle itself or
between the vehicles and the concerned infrastructure
as used in a network. However vehicular ad-hoc
network comes with several challenging characteristics
such as the potentially large scale and high mobility.
With the increase in distance the power or required
energy for the reliable and good communication is also
challenging task for the nodes in a network. The target
of vehicular ad-hoc network is achieving higher level of
road safety and decreases the number of accidents and
manages the overall traffic.
In previous work, they calculated two trusted nodes in
the cluster on the basis of highest energy. Malicious
node has also highest energy and selected as a trusted
node. Then it can forward the data to another attacker
and it affects the whole network. They find two nodes in
cluster with highest energy and make them cluster
head. This CH is only considered as trusted and all the
communication had done by these CHs only. When data
has to transfer then CH send packet to another cloud’s
CHs only. By this they eliminate the malicious nodes
from the network and make path more secure. In our
proposed work, we are finding only one cluster head
which is most trusted node in a cluster. So, we are using
transmission range with highest energy to calculated
cluster head. They have to fulfill both the criteria to
eliminate the selection of malicious node.
et

The rest of the paper is organized as follows section II
Discuss the related work, section III Discuss the
propose methodology and flow graph, section IV
Simulation setup V Discuss about Experimental and
finally conclusion and future work.
II. RELATED WORK
The most relevant issues in vehicular ad-hoc network is
to provide quality of services and the real time
intermittent connectivity between the vehicle nodes, and
the high dynamic mobility. The passive multi-hop
clustering algorithm is proposed [1] they have followed
the cluster head selection mechanism to crate the
cluster using passive multi-hop algorithm, here a priority
based neighbor following strategy is proposed and find
the optimal neighbor by calculating their parameters. A
stable routing algorithm proposed [2] author proposed
the cluster based mechanism to select the optimal route
stability and find the best quality of services in vehicular
ad-hoc network, the selected routing algorithm choose
the best stable and reliable path between the sender
and receiver in the vehicular ad-hoc network. The
classification of routing protocol proposed [3] here the
author done multi hop structure protocol using the
hierarchical, flat and location-based routing protocol,
with using all these protocol techniques they compare
the different number of parameters. The current
challenges and trends presented [4] in this paper author
review the different techniques and discuss the about
current trends and challenges in the present vehicular
ad-hoc network. Here they also present the different
hardware mechanism, spectrum, security and other
privacy issues. The mobility based and stability-based
clustering algorithm is proposed in [5], here the author
present clustering model for the especially urban area,
where they record each vehicle movement and position
with their direction and lifetime estimation. They also
calculated with their proposed techniques in average
number of cluster and the average number of cluster
head life time. The unified framework of clustering
approach presents here [6] author present some
important points such as the neighbor sampling of the
vehicle node, cluster head selection process based on
the back-off heads and at the last cluster maintenance
based on the backup cluster head process. Also define
the matrix for the vehicle node including some
parameters such as the vehicle mobility, vehicle position
and velocity with the vehicle node life time in a network.
An approach proposed by [13] was employed optimized
link state routing this will help us to understand unicast
routing protocol for adhoc network. Santa et al., has
been modified here to build a more robust method to
use in VANET [14]. In OLSR, a set of Multi Points Relay
(MPR) nodes is selected by every vehicle, which is used
to retransmit the packets. MPR nodes decrease the
overhead of flooding message. OLSR is a proactive
routing protocol that each node has a route to every
other node in the network. However, the overhead of
message for retaining the routes is high. Badis and
More recent work by Agha [15] have designed QOLSR
protocol for ad hoc wireless networks using OLSR. This
protocol considers the QoS constraints during the
selection of optimal paths. Due to the QOLSR chooses
the optimal paths in terms of bandwidth and delay and
ignores the mobility of vehicles; it is unable to deal with
VANETs. Then, most recent and advance theory
proposed by Otrok et al., [16] have introduced the QoS
Optimized Link State Routing (QoS-OLSR). This
protocol considers the QoS of the vehicles like residual
energy and bandwidth during the selection of cluster-
head and MRPs. However, it ignores the mobility of
vehicles for computing the QoS.
III. PROPOSED METHOD
Intelligent traffic system and monitoring is required for to
improve the current traffic monitoring system and road
safety features. In previous study author sole the
clustering issue using passive multi-hop clustering
techniques to improve reliability and stability of nodes in
vehicular ad-hoc network. In this section we present the
proposed methods for the vehicular ad-hoc network to
provide the reliable and stable communication between
the vehicle nodes, here we proposed approach i.e. multi
hop clustering algorithm using with the cluster head
selection process. In the below diagram we mention the
proposed MHC (Multi hop clustering Approach) flow
chart for the communication between the vehicle nodes.
Initially we start the simulator and send the hello
message to establish connection for all the participating
nodes in a network, after receiving the hello message all
nodes reply with their identification number, their
location, their distance between the nodes and the
destination. The cluster head node is selected if the list
is empty then road side unit select the any neighbor
node for join the cluster head. The cluster head node is
selected on the basis of closest speed of node with
average speed of node and having a more network life.
Then road side unit informs all the participating nodes to
elected as a cluster head node in a network then road
side unit find the gateway for the cluster head node and
share the information with the neighbor node form the
destination, and the finally process end. This paper also
calculates the distrust value of every vehicle to keep
record of white and black list of vehicle white vehicle are
honest vehicle and black are dishonest and malicious
which propagate the false message in the network lead
to misshape in road and traffic in VANET network. We
have added QoS matric to calculate the stability of the
network and path length.
A. Calculation of distrust Value
Due to lack of infrastructure and central management,
vehicular ad hoc networks are vulnerable to a number of
security threats. Vehicles can cooperate with each other
and increase security in the network. The following
method is utilized for verifying the behaviour of vehicles
and isolate malicious vehicles. Flowchart of this method
is shown in Fig. 2. To select a cluster-head, distrust
values of vehicles should be calculated. It is worth
mention that a cluster key is assigned to each cluster-
head, which is specified by its certificate authentication
(CA). CA acts as a trusted third party and manages
cryptography keys, personalities and
certificates of vehicles located in its region. The initial
distrust value that is allocated to vehicles when they join
to the network is the same for all vehicles. This initial
value is set to 1. At first, each vehicle is placed inside
the white list and then if its distrust value becomes
larger than a threshold value, it moved to the black list.
The threshold value δ is a criterion for updating black
and white lists [11]. It is computed in each environment
as follows:
δ = eφ 0 ≤ φ ≤ Kv-1
where Kv is the average number of vehicles in a certain
environment and is computed as:
‫ݒܭ‬ = [Navg / Ravg]
where Ravg and Navg are the typical transmission range
and the average number of vehicles, respectively.

Fig. 2. Proposed flowchart for the cluster head selection
process.
B. QoS value:
After calculation the distrust value, we offer to calculate
Qos model metric a vehicle can calculate its QoS value
using its neighborhood table and obtaining information
about the network. QoS value is calculated [11]
QoS =(Bv× Nv× Dratio/vratio)/Dv
Bv= Available bandwidth
Nv = number of neighbors of vehicle
D_ratio=ratio of Remaining Distance
V_ratio=Ratio of Velocity
After adding these two new concepts in two existing
work now we have calculated two new parameter which
stable our cluster and it’s also calculate the distrust
value for keep aside to malicious and dishonest from the
network. It is discussed below.
(a) Percentage of stability: The percentage of stability
can be calculated by dividing the number of current
vehicles in each cluster to the previous number of
vehicles in the same cluster before a slot of time.
Indeed, stability of cluster is the average lifetime of a
cluster in terms of the number of nodes within that
cluster. If over 60% of the nodes are in the cluster, the
cluster is intended stable.
(b) Path length: This criterion is described as the
average number of hops that used to transmit data
between the source node and destination node. The
path length can be reflected in the
end-to-end delay
IV. SIMULATION SETUP
The performance of our proposed approach is evaluated
using simulation implemented with NS-2.35 under Linux
Ubuntu 16.04. Ns2 is the most popular simulator for
academic research and itis more difficult than other
software. It is a discrete event and an object-oriented
simulator designed for networking research. Also, it
models different network architectures, including
Wireless LAN, MANET, VANET and satellite. NS2 is
written in C++ with an OTcl (Object Tool Command
Language) interpreter. Hence, the proposed approach
has been implemented by this programming language.
MOVE that is the generator of mobility model is used for
vehicular networks. It is based on the Java
programming language and builds on SUMO
(Simulation of urban mobility) that is a time discrete and
open-source microscopic road traffic simulation
package. This traffic simulator applies XML code to
represent the network features such as number of
vehicles, velocity, duration and topography. The area of
simulation selected is the highway of city. We use
SUMO 0.12.0 to generate the vehicle traffic and export
a part of city map from Open Street Map in the form of
XML formatted Osmfiles.
A simulation area of 3000 m × 1000 m is used to
simulate by varying the number of vehicles from 10 to
50. Also, the highway topology is applied to simulate the
traffic. We assume that all nodes generate a constant bit
rate (CBR) peer-to-peer data traffic.
Moreover, the vehicle that drops or duplicates packets
and changes the content of the packets is considered as
malicious vehicle. The percentage of the malicious
vehicles in the network is considered 10% of the whole
number of vehicles in the network. Also, we take a 95%
confidence interval to obtain more precise simulations.
The simulation parameters used for NS2 are given in
Table 1.
Table 1.
Parameter Value
Time of Simulation 400s, 600, 800, 1000s
Dimension 3000 ×1000
Traffic Model CBR(Constant Bit Rat)
Transmission range 250M
Data Packet Size 512byte
Topology High Way with Two bands
Protocol AODV
No of Vehicle 15,30,45,60,75
Vehicle Speed 60 to 120 Km/h
Idle time Random Value b/w (0,1)
Link band width
1 Mbps
Available Band width Idle time × Link bandwidth
MAC/PHY IEEE 802.11p
V. EXPERIMENTAL RESULT
In this section we discuss about our proposed
methodology result which is based on the existing
techniques and the proposed method i.e. multi-hop
clustering (MHC) techniques were simulated with the
network simulator 2.34. here the all calculated results
were mentioned in the below table and store the value
of each parameters and we added two more parameter
in the existing work which is percentage of stability and
path length calculation .percentage of stability shows
cluster head stability which gives the strength to network
so the calculated are network overhead, throughput,
energy consumption, end to end delay, packet loss,
percentage of stability and path length calculation .
Packet delivery ratio: the packet delivery ratio is
obtained by dividing the number of packets successfully
received by the number of packets originated for a
destination [12].
- End-to-end delay: end to end delay refers to the time
required to transmit a packet across a network from
source to destination [12].

Throughput: the throughput is calculated by dividing the
total number of data bits delivered to destination node
during the simulation by the total simulation time [12].
Number of packets loss: number of packet loss is
defined as the number of packets not received at their
destination during the simulation time [12].
Percentage of stability: the percentage of stability can
be calculated by dividing the number of current vehicles
in each cluster to the previous number of vehicles in the
same cluster before a slot of time. Indeed, stability of
cluster is the average lifetime of a cluster in terms of the
number of nodes within that cluster. If over 60% of the
nodes are in the cluster, the cluster is intended stable.
Path length: this criterion is described as the average
number of hops that used to transmit data between the
source node and destination node. The path length can
be reflected in the
We have discussed many parameters here and many of
them is discussed in previous work and two of them are
calculated in this paper and we have compared this
parameter with VANET QoS-OLSR, the cluster-based
QoS-OLSR and the classical QOLSR is illustrated them
are discusses.
The results were directly compared with previously
reported finding and compared between the proportion
of stability of MHC approach in VANET and different
protocols is shown in Fig 3. This figure represents that
in comparison to other protocol the percentage of
stability of MHC approach -VANET is increased by
approx. More than 10%. This is because our proposed
protocol uses the distrust value and the proportional
distance and velocity to calculate the QoS value per
vehicle. It guarantees that cluster-heads and gateways
are selected with proper velocity and significant distance
to traverse and avoids the repeated disconnections.
Fig. 3. Compression Percentage of Stability.
This result demonstrates the potential superiority
approach over previous establish methods Fig. 4
compares the number of packet loss obtained by
different protocols. Due to the proposed protocol
increases the percentage of packets transferred, the
number of packet loss of MHC approach VANET is
lower than other investigated protocols about 6%.
represents the average number of hops yielded using
the investigated protocols. This figure demonstrates that
the number of hops in MHC-approach is about 3% less
than other examined protocols. This is due to the fact
that this protocol calculates the QoS function and select
the nodes that has the highest value of QoS as
gateway.
Fig. 4. Compression of Path Length.
VI. CONCLUSION
The main conclusions of this work are drawn together
and presented in this section. This paper presents the
new multi-hop based advanced clustering techniques for
the efficient routing and stability of the nodes in a
network. The proposed techniques also used
optimization techniques for the various parameters to
record and calculate the value of node. Our simulated
result shows better results than previous techniques
with the parameters such as Percentage of stability and
Path length. In future research we use some
evolutionary techniques and some hybrid model for the
improvement in clustering issue and trust node-based
calculation in network.
VII. FUTURE WORK
In future work we may also use trust node-based
calculation for the authenticity of the node in the
vehicular ad-hoc network, and also used some
advanced classification techniques to enhance the
performance of the network.
ACKNOWLDEGEMENTS
I would to like to thank my supervisor Dr. Pushpneel
Verma and Dr. Jay Trilok Choudhary for his constant
help, support and encouragement without there
guidance, this work would impossible to complete. I
would also like to thank my family and friends for their
moral support.
REFERENCES
[1]. Zhang, D., Ge, H., Zhang, T., Cui, Y. Y., Liu, X., &
Mao, G. (2018). New multi-hop clustering algorithm for
vehicular ad hoc networks. IEEE Transactions on
Intelligent Transportation Systems, 20(4), 1517-1530.
[2]. Mu'azu, A. A., Tang, L. J., Hasbullah, H., & Lawal, I.
A. (2015). A cluster-based stable routing algorithm for
vehicular ad hoc network. In 2015 International
Symposium on Mathematical Sciences and Computing
Research (iSMSC), 286-291.
[3]. Rani, S., & Ahmed, S. H. (2016). Multi-hop Network
Structure Routing Protocols. In Multi-hop Routing in
Wireless Sensor Networks (pp. 45-58). Springer,
Singapore.
[4]. Rasheed, A., Gillani, S., Ajmal, S., & Qayyum, A.
(2017). Vehicular ad hoc network (VANET): A survey,
challenges, and applications. In Vehicular Ad-Hoc
Networks for Smart Cities (pp. 39-51). Springer,
Singapore.

[5]. Ren, M., Khoukhi, L., Labiod, H., Zhang, J., &
Veque, V. (2016). A new mobility-based clustering
algorithm for vehicular ad hoc networks (VANETs).
In NOMS 2016-2016 IEEE/IFIP Network Operations and
Management Symposium (pp. 1203-1208). IEEE.
[6]. Ren, M., Zhang, J., Khoukhi, L., Labiod, H., &
Vèque, V. (2017). A unified framework of clustering
approach in vehicular ad hoc networks. IEEE
Transactions on intelligent transportation
systems, 19(5), 1401-1414.
[7]. Rossi, G. V., Fan, Z., Chin, W. H., & Leung, K. K.
(2017). Stable clustering for ad-hoc vehicle networking.
In 2017 ieee wireless communications and networking
conference (wcnc) (pp. 1-6). IEEE.
[8]. Shanmugasundaram, G., Thiyagarajan, P., Tharani,
S., & Rajavandhini, R. (2017). A Multilevel Clustering
Using Multi-hop and Multihead in VANET.
In Proceedings of the International Conference on Data
Engineering and Communication Technology, 171-178.
[9]. Singh, A., & Kaur, M. (2015). A novel clustering
scheme in vehicular ad hoc network. International
Journal of Applied Information Systems (IJAIS), 10(3),
1-5.
[10]. Sugumar, R., Rengarajan, A., & Jayakumar, C.
(2018). Trust based authentication technique for cluster
based vehicular ad hoc networks (VANET). Wireless
Networks, 24(2), 373-382.
[11]. Fatemidokht, H., & Rafsanjani, M. K. (2020). QMM-
VANET: An efficient clustering algorithm based on QoS
and monitoring of malicious vehicles in vehicular ad hoc
networks. Journal of Systems and Software, 110561.
[12]. Ankit Temurnikar, Pushpneel Verma, Jay Trilok,
Choudhary. Securing Vehicular Adhoc Network Against
Malicious Vehicle Using Advanced Clustering
Technique. Paper Under Proceeding “ IDEA 19-20
IEEE conference RGPV Bhopal.
[13]. Clausen, T., Hansen, G., Christensen, L., &
Behrmann, G. (2001). The optimized link state routing
protocol, evaluation through experiments and
simulation. In IEEE symposium on wireless personal
mobile communications, 1-6
[14]. Santa, J., Tsukada, M., Ernst, T., Mehani, O., &
Gomez-Skarmeta, A. F. (2009). Assessment of VANET
multi-hop routing over an experimental
platform. International Journal of Internet Protocol
Technology, 4(3), 158-172.
[15]. Badis, H., & Agha, K. (2005). QOLSR, QoS routing
for ad hoc wireless networks using OLSR, European
Transactions on Telecommunications, 16, 427–442.
[16]. Otrok, H., Mourad, A., Robert, J. M., Moati, N., &
Sanadiki, H. (2011). A cluster-based model for QoS-
OLSR protocol. In 2011 7th International Wireless
Communications and Mobile Computing Conference,
1099-1104.
How to cite this article: Temurnikar, A., Verma, P. and Choudhary, J. (2020). Development of Multi-Hop Clustering
Approach for Vehicular Ad-hoc Network. International Journal on Emerging Technologies, 11(4): 173–177.

VANET Clustering

More Related Content

What's hot (19)

Similar to VANET Clustering (20)

Recently uploaded (20)

VANET Clustering