Content uploaded by Mustafa Maad Hamdi
Author content
All content in this area was uploaded by Mustafa Maad Hamdi on Nov 12, 2023
Content may be subject to copyright.
979-8-3503-4215-4/23/$31.00 ©2023 IEEE
Successful Delivery using Stable Multi-Hop
Clustering Protocol for Energy Efficient Highway
VANETs
Mustafa Maad Hamdi
College of Computer Science and IT,
University of Anbar,
Al-Anbar, Iraq.
mustafa.maad.hamdi@uoanbar.edu.iq
Salah Ayad Jassim
Renewable Energy Research
Center, University of Anbar,
Al-Anbar, Iraq.
salah.ayad@uoanbar.edu.iq
Baraa Saad Abdulhakeem
College of Computer Science and IT,
University of Anbar,
Al-Anbar, Iraq.
Baraasaad@uoanbar.edu.iq
Abstract—
Vehicular ad hoc network (VANET) is a highly
developing technology hence it is created using the moving vehicles it
occupies maximum of the attention of the research scholars and the
industrial people and as well it occupies most of the applications of
Intelligent Transmission System (ITS). Vehicles in the VANET
networks provide communication through two modes transmission, they
are; inter vehicular and infrastructure based communication. Due the
core functionalities of VANETs like high speed, random mobility and
unpredictable vehicles movements certain drawbacks are occurred in
the vehicular network which leads to increase of power utility, delay and
overhead. In recent times cluster based models becomes more popular
and as well it needs improvement to handle to high speed vehicles. For
that purpose in this article, Successful Delivery using Stable Multi-Hop
Clustering Protocol (SSMC) for VANETs are developed which includes
effective clustering formation and cluster head (CH) selection process,
cluster architecture and multi-hop communication among the vehicles.
With the presence of these processes the communication quality of the
vehicles can get increased. The implementation of this single-hop and
multi-hop communication in VANETs are carried out in NS2 were its
vehicle mobility is generated using the SUMO software. By calculating
certain parameters like energy efficiency, packet delivery ratio, end to
end delay and routing overhead the performance of the proposed
SSMC-VANETs are evaluated and it gets compared with the earlier
methods like MDEC-VANETs and COIM-VANETs. From the results
evaluation is it observed that the proposed SSMC-VANETs attain
maximum efficiency and delivery ratio when compared with the other
baseline methods.
Keywords— Vehicular ad hoc network (VANET), Intelligent
Transmission System (ITS), Multi-Hop Clustering, Cluster
Formation, Cluster Head Selection.
I. I
NTRODUCTION
VANETs network model is highly innovative as it is
attracted by researchers of several industrial areas and as well
lots of intensive research are developed in recent times.
VANETs are designed in the way to provide effectual
communication among the vehicles in an adhoc networks. The
communication entities which are present in the vehicles
architecture are Road Side Unit (RSU) and On-Board Unit
(OBU). OBU are equipped inside each vehicle which allows
performing wireless communication with the RSU stationary
unit. The communication among the vehicles is carried out in
two modes like in-between two vehicles directly and it
between the vehicle and infrastructure which is the RSU. The
standard and the wireless channel which is used allow inter
vehicular communication is IEEE 802.11p and Dedicated
Short-Range Communications (DSRC). The communication
model of VANETs is illustrated in Fig.1.
Fig. 1. VANETs Communication Model
The core characteristics of VANETs are high speed
communication, varying vehicle density, dynamic topography
among the vehicles and that leads to the increase of frequent
link failure at the time of data transmission among the vehicles
which increases the power utility, delay and overhead. The
standard possible solution to overcome this problem is
vehicles clustering. In general clustering is known as grouping
of vehicles into cluster head (CH) and its members. As the
results of this grouping process power utility among the
vehicles are reduced. But in recent time the density of vehicles
are huge and the standard clustering model is not applicable
and it needs improvement. For that purpose in this paper
successful delivery using stable multi-hop clustering protocol
and its contribution is discussed.
The contribution of this paper includes an effective cluster
formation and cluster head selection process to attain better
communication quality among the vehicles. Additionally, here
cluster architecture is improvised and the multi-hop
communication is concentrated so that it can able to attain
effective communication. The structure of the paper is listed
here; in section 2 the baseline researches in the area of
VANETs clustering is performed and its drawbacks are
identified. Section 3 the proposed SSMC-VANETs approach
is elaborated. The performance analysis and the results
calculations are given in the sections 4 and 5. Finally the
limitation and future direction is shown in the conclusion
section 6.
II.
RELATED WORKS
In [5], the author Shashank K. Gupta et.al, presents an LTE-
based architecture for vehicular networks, emphasizing
safety message transmission. It introduces two
communication protocols, CMMP and CMBMP, which
outperform existing techniques. These protocols offer low
warning message delivery delays and high delivery ratios in
hybrid LTE networks. In [6], the author Gurjot kaur et.al,
proposes a novel transmission approach that optimizes
2023 7th International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT) | 979-8-3503-4215-4/23/$31.00 ©2023 IEEE | DOI: 10.1109/ISMSIT58785.2023.10304927
Authorized licensed use limited to: UNIVERSITY TENAGA NASIONAL. Downloaded on November 10,2023 at 22:02:12 UTC from IEEE Xplore. Restrictions apply.
network topology based on received SINR by Connected
Vehicles (CVs). It successfully balances single-hop
performance and dual-hop resource utilization, achieving a
coverage probability of 0.66 with 80% resource usage,
compared to an ideal 0.84 in single-hop transmission. In [7]
the author Weijing Qi et.al, introduces a Traffic
Differentiated Clustering Routing (TDCR) mechanism for
hybrid vehicular networks, addressing data collection
challenges. TDCR optimizes data delivery in VANETs by
balancing cellular bandwidth costs and latency through a
centralized clustering approach.
In [8], the author Mohammed S. Bahbahani et.al, introduces
a novel approach for achieving ultra-reliable, low-latency,
and high-data-rate communication (URLLHDC) in vehicular
networks by integrating millimeter-wave (mmWave)
communication with the DSRC standard. The proposed
technique forms vehicle clusters with directional mmWave
antennas, enabling URLLDHC in various road conditions
with minimal overhead. In [9] the author Florian Klingler
et.al, introduces a novel scheme using Bloom filters to
maintain 2-hop neighbor data in dynamic wireless networks,
particularly focusing on VANETs. The proposed approach
decreases the length of beacon packets, minimizing channel
load and collision probability. In [10] the author Ning Liu
et.al, introduces the Global Topology-Based Broadcast
(GTB) algorithm for VANETs, utilizing base stations to
obtain global topology information and optimizing cluster
head selection. GTB demonstrates superior performance in
reliability and latency compared to existing broadcast
algorithms in both urban and freeway scenarios.
In [11] the author Celimuge Wu et.al, presents a vehicle to
RSU based communication protocol using distributed
clustering, coalitional game theory, fuzzy logic, and
reinforcement learning for optimized cluster formation and
route selection. It also includes a multi-hop data delivery
virtualization scheme. Simulations demonstrate the protocol's
superior performance compared to other approaches. In [12],
the author Xuming Zeng et.al, presents a probability-based
newer broadcast model for vehicles, which assigns higher
forwarding probabilities to distant nodes, employs clustering
to reduce redundancy and latency, and enhances reliability.
In [13], the author Dali Hu et.al, investigates outage
probability in a VANET with random multiple access,
considering path loss, interference, node mobility, and
transmission probability. It provides bounds and seeks to
optimize parameters for improved network reliability.
In [14], the author Gunasekaran, developed a newer model to
improve the delivery ratio and network throughput called
location aided routing approach which includes clustering
and multi-hop dissemination to attain effective
communication among the vehicles but however it fails to
reduce the communication delay during communication
among the vehicles with high density. In [15], the author
Esha, proposed a novel method to improve the network
stability which includes multi hop communication, Internet
connectivity and constant verification of network traffic but
this model consume more power during communication
which affects the reliability of the network. In order to
overcome the earlier network drawbacks SSMC-VANETs
approach is developed and it is elaborated in the upcoming
section.
III.
PROPOSED SSMC
-
VANET APPROACH
In this study, we assume that each vehicle has an On Board
Unit (OBU) to support the IEEE802.11p as a Dedicated Short-
Range Communications (DSRC) system. By transmitting
Cooperative Awareness Messages (CAM), a single-hop
broadcast communication, the vehicles can periodically share
their data. A concrete heartbeat rate message is used to
regularly send this CAM message. Each car will be aware of
its close neighbors so that effective communication can be
attained among the vehicles. The major sections of these
proposed SSMC-VANETs are Cluster formation, CH
selection, Multi-hop clustered architecture and multi-hop
routing model. The structure of the proposed work is
illustrated in Fig. 2.
Fig. 2. Structure of the SSMC-VANETs
A. Cluster Formation
Cluster setup starts at the point when the vehicles count on
the highway rises, causing every vehicle which doesn't have
enough space or that needs to know the state of the roads to
activate the SSMC-VANETs algorithm based on the
information collected from neighbor vehicles. If there is
enough vehicles close enough to one another to create a group
called cluster with regard to and MinPts, SSMC-VANETs
analyzes the number of vehicles to make this determination.
The minimal number of points that are classified as neighbors
for a point is represented by MinPts. The average time which
is taken for execution for a query in terms of SSMC-VANET
complexity is O(log n).
B. CH Selection
Cluster stability, one of the performance requirements in
VANETs, is greatly influenced by Cluster Head Selection
(CHS), which is also important. More stable clusters result
from fewer CH changes. Following cluster establishment, the
CHS process begins with the election of one of the cluster's
cars as a CH. Finding the best CH in the cluster is done using
the Fuzzy Control (FC) method. The speed of cluster (SC),
Vehicle Acceleration (VA), and Confidence Supremacy (CS)
are the three factors taken into account throughout the CHS
phase. Calculating the average speed of the vehicles in the
clusters yields the SC value. The distance from the cluster
center is indicated by the letter CS. As a result, in any cluster
a vehicle is closer to all other vehicles the more central it is.
Each vehicle in the cluster's SC metric is determined by
CCx
∑,
(1)
These parameters are fuzzyfied using the FC method, where
N implies the cluster vehicles count and d(x,y) implies the
distance among the vehicle (x) to the other ones which is
present inside any cluster. The vehicle with the highest CHS
Authorized licensed use limited to: UNIVERSITY TENAGA NASIONAL. Downloaded on November 10,2023 at 22:02:12 UTC from IEEE Xplore. Restrictions apply.
measure will proclaim itself to be a cluster head after
calculating the CHS values for each vehicle in the cluster.
C. Multi-Hop Cluster Architecture:
The priority neighbor following based multi-hop cluster
model is used here, where Cluster A resembles a conventional
single-hop communication model inside the clusters with the
presence of less number of neighbors. Through the multi-hop
based communication process cluster B is created were the
CH is allocated for the vehicles 6 and there is two neighbor
vehicles are present among the vehicle 6 CH and the vehicle
11. We suppose that each and every vehicle is outfitted with
an on-board device which helps to perform long distance
communication among the vehicles in the network. The full
network consists of the normal vehicles in the cluster is
known as cluster member (CM) and cluster head (CH) state,
may only connect with the CM or CH nodes through the
WAVE protocol and cannot directly communicate with the
roadside unit. In addition to communicating with CM nodes
through WAVE, the vehicle node in CH mode is also capable
of communicating over a 4G network with RSU equipment.
The sole channel of communication that the CM node has
with the RSU under this communication architecture is CH.
Each vehicle stores the data in the routing table
(ROUT_TABLE) that provides data on the movements of the
vehicles and here the communication is predefined and fixed
number of maximum hop vehicles (MAX_HOP).
D. Multi-Hop Broadcast Protocol
The suggested approach is implementable as a multi-hop
broadcast protocol in the real time environment. As an
overview, the deployment process is as follows. The protocol
may be included in the functional module which handles both
the vehicles and RSU based communications in the Vehicle
Communication System (VCS). Once the car reaches a
highway system, the protocol in the VCS's application layer
will turn on. After discovering the topology, at the earlier
stage to transfer the data from one place to another the
vehicles utilizes the existing CH.
If sufficient cars have left or entered the network, it could also
start a clustering process. Second, during clustering, each
vehicle creates a hop based routing table that contains the data
transfer details of CHs to that particular vehicle. A node's
relative distance from the transmission border determines its
index in the neighbor table; a closer distance is represented
by a lower index number. Third, the vehicle that
unexpectedly runs into a problem becomes the broadcast's
source node. As soon as possible, it must broadcast an urgent
message or packet to every vehicle while switching the traffic
flow pattern to current direction to other.
Each vehicle may identify if a broadcast packet it receives is
from sender vehicles to perform the data forwarding process
by adhering to the protocol by comparing the GPS
coordinates and the direction of the traffic. We consider the
network of vehicles as a whole to be fully linked. The
broadcast packet can be successfully received by every
vehicle which is present in its coverage area.
1) Sender vehicle S transmits an emergency packet including
its vehicle identification details, GPS, and hop based routing
table to each vehicles with gets inside its coverage area. The
emergency packet will also be kept in its buffers for
retransmission at the end of the predefined transmission time
period.
2) When the emergency packet has been broadcast, sender
vehicle S begins is data transmission with a presence of a
down-counter. This down-counter's starting value reflects the
determined expiry time for this single hop data transmission
which is specified in the network. In case, if no nodes within
the sender vehicle coverage area have forwarded the original
broadcast packet by the time this counter gets reduced to ‘0’,
at the time the sender vehicle will retransmit the data with the
presence of its buffer space.
3) Those who can forward are just CHs. Each CH determines
the forwarding probability on its own. Upon receipt of the
urgent packet when gets received from sender vehicle S
which creates a chance at random among 0’s and 1’s. It
transforms into the forwarder, or the sender vehicle to its
nearby neighbor, and then transmits the emergency packet. If
the currently obtained forwarding probability is high when
compared with the generated random probability values that
leads to the creation of newer event ID
4) To attain effective communication among the vehicles
the CSMA/CA mechanism is included with the MAC layer so
that the data forwarding process becomes more effective
according to the step 3 procedure.
IV. SIMULATION EXPERIMENTS
In the section that follows, the SSMC-VANETs
simulation result estimate strategy is examined and compared
with well-known methods like the MDEC-VANETs [14] and
the COIM-VANETs [15]. The NS2 simulation application,
which runs on the Ubuntu operating system, is used to run the
simulation. By taking into consideration energy efficiency,
packet delivery ratio, end to end delay and routing overhead,
the findings are derived. The input research parameters which
are used for the implementation of this vehicular network are
described in Table 1.
TABLE I. PARAMETER SETTING
Parameters Values
Software Utility NS2.35 and SUMO
Simulation Run Time 150 ms
Network Coverage 1500m*1500m
No of vehicles 250 vehicles
Queue Type DropTail
Initial Energy 1000 mJ
Data Rate 250 Kbps
Data Traffic CBR
A. Energy Efficiency Calculation:
It refers to the process of quantifying how effectively the
sensor nodes within a network utilize their limited energy
resources to perform communication tasks, especially in
scenarios where the number of nodes can vary. When
compared to existing methods like MDEC-VANETs and
COIM-VANETs, the Proposed SSMC-VANETs achieves
superior efficiency, as shown in figure 3 and it gets attain with
the presence of effective cluster and multi hop
communication process among the vehicles in the network.
Authorized licensed use limited to: UNIVERSITY TENAGA NASIONAL. Downloaded on November 10,2023 at 22:02:12 UTC from IEEE Xplore. Restrictions apply.
Fig. 3. Energy Efficiency
B. Packet Delivery Ratio Calculation:
It represents the ratio of information which gets generated at
the source vehicle to the ones that effectively reaches the
destination vehicle. Figure 4 suggests the graphical
representation of the delivery ratio calculation, and it is clear
from that that the proposed SSMC-VANETs outperform
existing techniques like MDEC-VANETs and COIM-
VANETs in terms of performance. With the presence of
effective clustering and routing process the vehicles are
grouped in an effectual manner which helps to attain
maximum delivery ratio in the proposed method.
Fig. 4. Packet Delivery Ratio
C. Routing Overhead Calculation:
It involves tallying the overall count of information produced
by the source and the total count of data packets dispatched
to all nodes. When compared to existing methods like
MDEC-VANETs and COIM-VANETs, the proposed SSMC-
VANETs provides reduced routing overhead, as shown by
Figure 5's graphical explanation of routing overhead
calculation. It gets attain using the clustering process
involved in the proposed model. Such clustering process
reduces the data forwarding ratio among the vehicles and that
leads to the reduction of the routing overhead at the time
execution.
Fig. 5. Routing Overhead
D. End to End Delay Calculation:
It involves determining the duration required by a vehicle to
generate information and then transfer that to the required
destination. Figure 6 shows such delay calculations for the
techniques utilized for this research with varied numbers of
vehicles from 25 to 250 in counts, and it is clear from that that
the Proposed SSMC-VANETs created minimal end-to-end
delays when compared to the prior approaches like MDEC-
VANETs and COIM-VANETs.
Fig. 6. End to End Delay
V. RESULTS AND DISCUSSION
Here it examines the measurements of MDEC-VANETs and
COIM-VANETs and the proposed SSMC-VANETs strategy
with regard to packet delivery ratio, end-to-end delay, routing
overhead, and energy efficiency. The analyzed values which
are obtained from the NS2 implementation is described in
Table 2.
TABLE II. OBTAINED RESULTS AND ITS MEASURES
Parameters
/ Methods
MDEC-VANETs
COIM-VANETs
Proposed SSMC-
VANETs
Delivery
Ratio
79.26% 82.42% 91.46%
End to End
Delay
156.28ms 132.74ms 96.14ms
Energy
Efficiency
76.25% 81.46% 85.17%
Routing
Overhead
235 pkts 196 pkts 134 pkts
Authorized licensed use limited to: UNIVERSITY TENAGA NASIONAL. Downloaded on November 10,2023 at 22:02:12 UTC from IEEE Xplore. Restrictions apply.
Firstly, in terms of Delivery Ratio, the Proposed SSMC-
VANETs stood out by achieving an impressive 91.46%,
surpassing MDEC-VANETs by approximately 12% and
COIM-VANETs by about 9%, shows its superior data packet
delivery capabilities. Secondly, in End-to-End Delay,
Proposed SSMC-VANETs demonstrated the lowest delay at
96.14 ms, outperforming MDEC-VANETs by roughly 60 ms
and COIM-VANETs by around 37 ms, indicating quicker
data packet delivery. Thirdly, in terms of Routing Overhead,
Proposed SSMC-VANETs maintained the lowest overhead
with 134 additional packets, significantly reducing control
packet requirements compared to MDEC-VANETs (reduced
by 101 packets) and COIM-VANETs (reduced by 62
packets). Finally, concerning Energy Efficiency, Proposed
SSMC-VANETs excelled with an efficiency rate of 85.17%,
surpassing MDEC-VANETs by approximately 9% and
COIM-VANETs by about 4.7%, showcasing its ability to
utilize energy resources effectively for data transmission.
VI. CONCLUSION
This article is mainly developed to reduce the power
utilization and delay among the vehicles in the high speed
vehicular network. The core idea of the proposed SSMC-
VANETs includes cluster formation, CH election and its
architecture with multi hop communication. Such kind of
clustering process includes certain parameters like speed,
acceleration, fuzzy control method and confidence
supremacy to select the CH among the vehicles in the certain
regions and that greatly helps to reduce the power utility
which leads to increase of the vehicles efficiency. Through
multi hop communication process the CSMA/CA mechanism
is incorporated with the MAC layer that results in the
reduction of delay generation among the vehicles. The
experimental demonstration is performed in an open source
software called as network simulator 2 (NS2) with SUMO
includes to generate the high speed vehicles mobility. The
output parameters mentioned here to analysis the vehicles
performance are energy efficiency, packet delivery ratio, end
to end delay and routing overhead and as well the calculated
results are compared with the earlier baseline methods like
MDEC-VANETs and COIM-VANETs. From the
measurements it is proven that the SSMC-VANETs attains
9% to 12% high delivery ratio, 37 ms to 60 ms lower end to
end delay, 62 packets to 100 packets lower routing overhead
and 4% to 9% higher energy efficiency when compared with
the baseline methods. Still is consists of certain limitations
like it can able to attain better performance only with 250
vehicles maximum. So in future increasing the vehicles
density is the main concentration and research scope.
REFERENCES
[1] H. Alabbas and Á. Huszák, "CAMVC: Stable Clustering Algorithm for
Efficient Multi-hop Vehicular Communication on Highways," in 2020
43rd International Conference on Telecommunications and Signal
Processing (TSP), pp. 149-152, 2020, doi: 978-1-7281-6376-5/20.
[2] G. Khayat, C.X. Mavromoustakis, G. Mastorakis, J.M. Batalla, H.
Maalouf, S. Mumtaz, and E. Pallis, "Successful delivery in VANETs
with damaged infrastructures based on double cluster head selection,"
in 2020 IEEE 25th International Workshop on Computer Aided
Modeling and Design of Communication Links and Networks
(CAMAD), pp. 1-5, 2020, doi: 978-1-7281-6339-0/20.
[3] F. Yang, J. Han, X. Ding, Z. Wei, and X. Bi, "Spectral efficiency
optimization and interference management for multi-hop D2D
communications in VANETs," IEEE Transactions on Vehicular
Technology, vol. 69, no. 6, pp. 6422-6436, 2020, doi:
10.1109/TVT.2020.2987526.
[4] D. Zhang, H. Ge, T. Zhang, Y.Y. Cui, X. Liu, and G. Mao, "New multi-
hop clustering algorithm for vehicular ad hoc networks," IEEE
Transactions on Intelligent Transportation Systems, vol. 20, no. 4, pp.
1517-1530, 2018, doi: 10.1109/TITS.2018.2853165.
[5] S.K. Gupta, J.Y. Khan, and D.T. Ngo, "Clustered multicast protocols
for warning message transmissions in a VANET," in 2019 IEEE
Vehicular Networking Conference (VNC), pp. 1-8, 2019, doi: 978-1-
7281-4571-6/19.
[6] G. Kaur, D. Kakkar, and R. Singh, "A Novel Approach for
Connectivity Improvement in Cluster Based VANETs," in 2019 13th
International Conference on Sensing Technology (ICST), pp. 1-5,
2019, doi: 978-1-7281-4807-6/19.
[7] W. Qi, B. Landfeldt, Q. Song, L. Guo, and A. Jamalipour, "Traffic
differentiated clustering routing in DSRC and C-V2X hybrid vehicular
networks," IEEE Transactions on Vehicular Technology, vol. 69, no.
7, pp. 7723-7734, 2020, doi: 10.1109/TVT.2020.2990174.
[8] M.S. Bahbahani and E. Alsusa, "A directional clustering protocol for
millimeter wave vehicular Ad hoc networks," in 2020 IEEE 91st
Vehicular Technology Conference (VTC2020-Spring), pp. 1-6, 2020,
doi: 978-1-7281-5207-3/20.
[9] F. Klingler, R. Cohen, C. Sommer, and F. Dressler, "Bloom hopping:
Bloom filter based 2-hop neighbor management in VANETs," IEEE
Transactions on Mobile Computing, vol. 18, no. 3, pp. 534-545, 2019,
doi: 10.1109/TMC.2018.2840123.
[10] N. Liu, Y. Wu, Z. He, K. Niu, and C. Dong, "A Global Topology Based
Broadcast Algorithm for VANETs," in 2018 15th International
Symposium on Wireless Communication Systems (ISWCS), pp. 1-5,
2018, doi: 978-1-5386-5005-9/18.
[11] C. Wu, T. Yoshinaga, Y. Ji, and Y. Zhang, "Computational intelligence
inspired data delivery for vehicle-to-roadside communications," IEEE
Transactions on Vehicular Technology, vol. 67, no. 12, pp. 12038-
12048, 2018, doi: 10.1109/TVT.2018.2871606.
[12] X. Zeng, M. Yu, and D. Wang, "A new probabilistic multi-hop
broadcast protocol for vehicular networks," IEEE Transactions on
Vehicular Technology, vol. 67, no. 12, pp. 12165-12176, 2018, doi:
10.1109/TVT.2018.2872998.
[13] D. Hu, J. Wu, and P. Fan, "On the outage probability of interference-
limited multi-hop linear vehicular ad-hoc network," IEEE Access, vol.
6, pp. 75395-75406, 2018, doi: 10.1109/ACCESS.2018.2883648.
[14] Reddy B and M. Gunasekaran, “Dissemination and Multi-hop
Dissemination Based-on LoRaWAN for Emergency Communication
in Low Traffic Vehicular Ad Hoc Networks”, International
Conference on Intelligent Engineering and Management (ICIEM),
2022, doi: 10.1109/ICIEM54221.2022.9853130.
[15] Esha Agarwal and Deepti Kakkar , “Connectivity Improvement
in Cluster-Based VANET”, International Conference on Intelligent
Technologies (CONIT), 2022, doi:
10.1109/CONIT55038.2022.9848190.
Authorized licensed use limited to: UNIVERSITY TENAGA NASIONAL. Downloaded on November 10,2023 at 22:02:12 UTC from IEEE Xplore. Restrictions apply.
