Available via license: CC BY 4.0
Content may be subject to copyright.
VOLUME XX, 2017 1
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.Doi Number
Autonomous Vehicles with a 6G-based
Intelligent Cybersecurity Model
Abdullah Algarni1 and Vijey Thayananthan1
1Computer Science Department, King Abdulaziz University, Jeddah 21589, Saudi Arabia
Corresponding author: Abdullah Algarni (e-mail: amsalgarni@kau.edu.sa).
This work was supported by the Deanship of Scientific Research at King Abdulaziz University, Saudi Arabia, under Grant G-601-611-39.
ABSTRACT Sixth-generation (6G)-based communications have many applications and are emerging as a
new system to utilize existing vehicles and communication devices in autonomous vehicles (AVs). Electric
vehicles and AVs not supporting the integration of intelligent cybersecurity will become vulnerable, and their
internal functions, features, and devices providing services will be damaged. This paper presents an intelligent
cybersecurity model integrating intelligent features according to the emerging 6G-based technology based on
evolving cyberattacks. The model’s novel design was developed using the necessary algorithms to provide
quick and proactive decisions with intelligent cybersecurity based on 6G (IC6G) policies when AVs face
cyberattacks. In this model, network security algorithms incorporating intelligent techniques are developed
using applied cryptography. Money transaction handling services implemented in an AV are considered an
example to determine the security and intelligence level depending on the IC6G policies. Intelligence,
complexity, and energy efficiency (EE) are assessed. Finally, we conclude that the model results are effective
for intelligently detecting and preventing cyberattacks on AVs.
INDEX TERMS 6G security, autonomous vehicles, cybersecurity attacks, intelligent transportation system,
risk assessment
I. INTRODUCTION
All future systems will be automated with intelligent
connections; they will dominate all possible services and
actions quickly, efficiently, and intelligently. Based on the
current perspective in terms of intelligent cybersecurity, the
demand for smart and intelligent feature enhancement is
growing and becoming a prime concern, especially in terms
of achieving maximum security with a minimum associated
cost. Intelligent features aid Autonomous Vehicles (AVs)
when it comes to the proper maintenance of a vehicle’s
vulnerable parts, and also with situations regarding reckless
driving, severe accidents, lack of instructive driving, and
improper decisions, which incur extra expenses for
maintenance besides hindering national economic growth.
In AVs, features are added to activate autonomous
functions responsible for the internal electronic devices
controlling vehicle movements maneuvering, and operation.
These services are affected and damage the devices when
facing attacks, threats, unintelligent policies, and functional
errors because some functions are connected to external
services linked with external communication devices, such as
sensors. Intelligent cybersecurity is an essential solution that
intelligently and proactively solves many problems to secure
services internally and proactively.
All AVs have insurance policies that cover usage costs
associated with general wear and tear and also cover the
intelligent features of these vehicles. However, there are
limitations to intelligent cybersecurity based on 6G (IC6G)
policies, arising from the fact that these policies must be
created by intelligent experts who understand the 6G-based
intelligent systems and their security issues. According to [1],
the policy pathway to achieve a long-term vision reveals the
details of using AVs in the future. Policy packages towards the
superblock vision contain 6 themes that provide the necessary
processes to improve the overall transportation regulations in
the 2050 visions. Encouraging the sustainable adoption of
autonomous vehicles and policies for public transport in
Western countries [2] will increase economic benefits with
affordable security and safety. In [1-6], policies have been
introduced to improve security and safety in many
applications related to our research (AVs and 6G-based
systems). Regarding the limitations of the IC6G policies, we
must understand the licensed details of the final official release
of 6G.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
In this paper, we emphasize that the policies implemented
are directly proportional to the intelligence level of the
vehicle since all policies implemented affect the
vulnerabilities of the devices used in AVs. Furthermore, the
features of these AV devices should be governed by
operational policies with practical limitations. Taking this
into account, this work has the following objectives: (1) build
an intelligent cybersecurity model that influences the
policies implemented in AVs and in their devices, (2) secure
the services of all devices integrated within AVs, and (3)
improve the reliability of the devices, which would avoid
unnecessary vulnerabilities.
The heavy use of AVs has influenced the development of
many internal and external devices, such as sensors. With the
increase in IC6G users, there comes an increase in the
connectivity as well as the vulnerability and mobility of the
devices integrated within AVs; this in turn motivates many
interactions, increasing the number of unsecured
communications. The motivation for this research is to
minimize these issues, including overall energy consumption
and cost.
When accurate policies are not delivered on time,
vulnerabilities increase and the number of hacks made on a
system will grow, leaving the system defenseless. For
instance, decision-makers of intelligent banking systems
must be able to authorize and activate the appropriate
policies on time. The intelligent decision-makers of those
systems must deliver the policies in a timely manner;
otherwise, the services delivered by the banks will be
attacked. Here, on-time means that all factors should be
considered, taking into account the clients, servers, and all
interfacing links and communication.
Researchers have focused on intelligent transportation
systems (ITSs) using emerging technologies, including 6G.
Some recommended policies are also considered to investigate
the security of the services integrated within existing vehicles
and AVs. ITS was developed with many policies to improve
the safety of vehicles and maintain the regulations of
transportation services. In recent papers, intelligent
cybersecurity was also discussed with ITS to enhance security
solutions of transportation services.
In our proposed approach, we used policies based on the
IC6G policies. Intelligent cybersecurity focuses on improving
cybersecurity solutions of AV services influenced by policies
based on 6G requirements and intelligence levels, which are
proportional to the strength of the policies. For instance, when
the strength of the policies increases, the intelligence level in
intelligent cybersecurity solutions increases as well.
This paper makes the following contributions:
1) Portrays an overview of IC6G and its associated
emerging technology in autonomous vehicles,
2) Proposes a taxonomy for IC6G through an extensive
literature investigation,
3) Presents a conceptual model for IC6G to motivate
future researchers to enhance the level of security
solutions in AVs with strength of the policies, advanced
integrated devices and technology,
4) Presents a set of challenges and open research issues for
the discussion of novel ideas among researchers aimed at
enhancing the functions of IC6G, such as the levels of
cybersecurity solutions.
The rest of this paper presents a scheme for managing
traffic in the following sections: a literature review and
related works are presented in Section 2. Following that,
Section 3 presents the proposed research, which involves the
design of cybersecurity solutions, the 6G-based architecture
of the proposed model, and the intelligent features necessary
for autonomous vehicles. Section 4 shows the relevant
comparison tables and results that support this research as
outlined in the contributions list. We discuss the security
issues involved in AVs and intelligent management issues in
Section 5. Intelligent features influenced by policies and
their management issues in 6G are considered in Section 6,
in which a simple scenario shows the vulnerabilities and
cyberattacks that can occur from poorly maintained policies.
This section also includes the latest challenges and
limitations facing intelligent security management. Finally,
in Section 7, we provide conclusions and consider future
work involving the development of AVs with intelligent,
human-like vision.
II. LITERATURE REVIEW
In an AV with 6G-based intelligent systems and efficient
cybersecurity, connectivity is an essential technical concept
for improving secure services and infrastructure. Studying 6G
networks provides the best intelligent cybersecurity security
solutions.
Fig. 1 shows the road map toward 6G-based application
scenarios, displaying the key performance areas of future
intelligent services and the challenges in 6G networks that are
FIGURE 1.
The road map towards URLLC in 6G networks [7].
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
part of this research’s objectives. Although Enhanced Mobile
Broadband (eMBB) and massive Machine Type
Communications (mMTC) are important for improving
services, Ultra-Reliable Low Latency Communications
(URLLC) dominate 6G networks for knowledge-based
analysis and optimization, knowledge-assisted training of
deep learning, and fine-tuning of deep learning networks.
URLLC will definitely improve services due to its low energy
consumption, which will also reduce the cost of security
solutions.
The authors in [8] offer a security assessment for the
evolution of Vehicle-to-Everything Communications (V2X-
C) architecture and the integration of 5G and 6G networks.
They also provide a comparison of the Quality of Service
(QoS) versus security provisions for Connected and AV
(CAVs) and illustrate the safety and security enhancement
mechanisms for V2X-C. A deep CNN-LSTM architecture is
proposed in [9] for CAV intelligence threats and compared
with other deep learning algorithms such as DNN, CNN, and
LSTM.
Paper [3] proposed a System Dynamic model based on a
Causal Loop Diagram that integrated the main
interdisciplinary variables and evaluated the impact of the
Regulation and Policy Framework (R&PF) on CAVs’
cybersecurity by focusing on several aspects, such as the
constraints on privacy and data accessibility.
A security model proposed in [10] for 5G satellite-
connected Unmanned Aerial Vehicle (UAV) networks aims to
make communication more secure, as UAVs have recently
become targets for cyberattacks due to an increase in volume
and low information security levels. In addition, [10] states
that a huge number of UAV connections in the future will not
only use 5G or 6G but will also use communication network
technologies that are even more advanced. By optimizing
leveraging a particle swarm [11] proposed two attacks
(poisoning and evasion) versus traffic sign recognition
systems in AVs based on which phase of the machine learning
process is targeted during an attack.
The authors in [12] presented their perspective on an
advanced and autonomous UAV traffic management (UTM)
system enabled by 6G communication technology that uses
non-terrestrial networks (NTNs) to improve air transportation
management in terms of safety and efficiency. For a robust
system, [13] uses efficient communication resources and
privacy preservation learning to build a Dispersed Federated
Learning (DFL) framework for 6G-enabled autonomous
driving cars.
The authors in [14] also proposed 6G architecture as an
integrated system, enabling technologies to provide security
and intelligence. They also discussed core services, KPI, the
possible technical challenges of 6G, and potential solutions.
The authors in [15] give an overview of how Artificial
Intelligence (AI) can solve the challenges of security and
privacy of 6G networks, giving suggestions for possible
solutions. A model was proposed in [16] for malicious traffic
detection within 6G to develop efficiency and security at the
same time.
TABLE I
COMPARISON OF 6G-BASED INTELLIGENT SECURITY ISSUES IN THE AV NETWORK
Ref.
Security Issues
Mechanism
Description
[17]
Untrusted
communication
in connected
and
autonomous
vehicles
A trusted
autonomous vehicle
routing protocol.
The mechanism presents
an efficient and trusted
autonomous-vehicle-
routing protocol using
6G networks.
[4]
Cyber attacks
A multi-agent
reinforcement
learning algorithm
with a hybrid deep-
anomaly detection.
The mechanism is used
for autonomous vehicles
in a 6G-V2X
environment.
[18]
Cyber threats
A system-dynamics
model with six
approaches: i)
CAVs
communication
framework, ii)
secured physical
access, iii) human
factors, iv) CAVs
penetration, v)
regulatory laws and
policy framework,
and iv) trust—
across the CAVs-
industry and among
the public.
A conceptual system
dynamics model for
cybersecurity assessment
of connected and
autonomous vehicles.
[6]
Untrusted
environment
and network
components in
AVs
Intelligent zero-
trust (ZT)
architecture and
dynamic-trust
algorithm
Introducing key ZT
principles as real-time
monitoring of the
security state of network
assets and intelligent
zero-trust architecture for
5G/6G networks with
machine learning utilized
in the open-radio access
network (O-RAN)
architecture
[19]
Ransomware
attack
Deep-learning-
based novel
ransomware
detection
framework
Used to secure the
supervisory control and
data acquisition
(SCADA) in electric
vehicle charging stations
from ransomware
attackers
Many researchers have surveyed the relevant security
threats, issues, technologies, techniques, and solutions based
on the future use of 6G (Tables I and II). Other researchers
have also surveyed several Machine Learning techniques that
have been applied to vehicular communication networks,
especially in terms of security, and have forecast how
Artificial Intelligent (AI) will be integrated into 6G vehicular
networks [23, 24].
According to the findings in [25], autonomous vehicle
vulnerabilities may jeopardize autonomous services.
Consequently, researchers have identified various types of
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
autonomous vehicle attacks and their countermeasures. The
authors proposed three types of attacks: autonomous control
systems, autonomous driving system components, and
vehicle-to-everything communications. The authors of [26]
provide not only a comprehensive survey of cybersecurity but
also current countermeasure strategies for securing AVs and
their services.
Another study found that the four dimensions of
autonomous driving security are sensors, operating systems,
control systems, and vehicle-to-everything communication
[27]. [28] described AV attack models and countermeasures
for electronic control units (ECUs), sensors, intra-vehicular
links, and inter-vehicular links.
[29] provided specific details of autonomous systems to aid
the development of future autonomous-mobility services.
CAVs are vehicles outfitted with various internet-of-things
(IoT) sensors that collect security and safety data from their
surroundings. In [30], a new model for developing
autonomous services is presented. The authors identified
hedonic motivation, trust in AVs, and social influence on
security issues as significant factors in performance
expectations. Hedonic motivation is used to increase travelers'
trust in automated vehicles.
A previous study [1] established a security policy pathway
for the future use of AVs. Six themes are detailed in policy
packages aimed at the superblock vision and the processes
required to improve the overall transportation regulations
described in the vision for 2050. The study [31] concentrated
on the integration of intelligent transportation systems (ITS)
and AV with maximum security and safety.
By 2030, cybersecurity technology for selected security
issues (CVs and data communication countermeasures) for
autonomous-transportation services can overcome several
challenges using four countermeasures: AI-supplemented, AI-
generated, AI-mediated, and AI-facilitated. AI will dominate
CVs and data communication countermeasures in the next
TABLE II
COMPARISONS OF EXISTING/RELATED TECHNIQUES WITH THE APPROPRIATE PARAMETERS/VARIABLES
Ref.
Existing/Related Techniques
Parameters/Variables
[8]
Network function virtualization (NFV) and cloud techniques
Using CAVs communication parameters, seven facets of security
and safety (i.e., availability, authentication, reliability,
confidentiality, integrity, robustness, and trustworthiness) for a
smooth ITS operation are considered.
[9]
Deep CNN-LSTM architecture for CAV threat intelligence
assessed and compared the performance of the proposed
model against other deep learning algorithms, such as DNN,
CNN, and LSTM.
CAV-KDD dataset, input & control data, and parameters of CAV
threat landscape and intelligence are used.
[3]
Causal loop diagram-based system
dynamic model is considered a technique.
Critical interdisciplinary parameters are incorporated with this
model for analyzing cybersecurity.
[11]
Poisoning attacks with particle swarm optimization (PAPSO)
and evasion attack with particle swarm optimization (EAPSO)
are proposed as techniques.
Attacker’s goal, knowledge, and capacity are used as parameters.
[13]
Block successive upper bound minimization (BSUM)-based
solution proposed a technique supporting the dispersed
federated learning (DFL) framework for Avs.
Simulation parameters, such as vehicular network area, number
of Avs, and cellular users, are being focused on.
[14]
Emerging techniques of the 6G network are focused on 6G
core services influencing intelligence.
Parameters of security issues and other 6G requirements are used
to analyze the security performance and intelligence with policies.
[20]
Novel loss based on feature mapping and joint optimization
network techniques is used.
Parameters, such as the camera internal, different channels, and
affine transformation, are used for improving the intelligence of
the Avs.
[21]
Many use cases are discussed as techniques supporting the
enhancement of the research objectives, including intelligent
security and optimal resource allocation policy.
Security, privacy, energy efficiency, and intelligent traffic are
considered.
[22]
Federated learning (FL) of explainable artificial intelligence
(XAI) models
Signal-to-interference plus noise ratio (SINR) value measured at
a packet reception percentage at the frame arriving
simultaneously with its display.
[4]
Hybrid deep anomaly detection (HDAD)
Multi-agent reinforcement learning (MARL) algorithm
Maximum entropy inverse reinforcement learning (MaxEn-
tIRL).
Policy calculation with frequency rate, previous attack data,
current attack data, and QoS parameters.
[6]
Intelligent zero trust architecture for 5G/6G networks is
considered with the machine learning and RL algorithms.
Intelligent and dynamic policies are measured with security and
network traffic parameters.
[19]
Novel deep learning-based ransomware detection framework
with policies and regulations is used as a technique.
Layer1-Layer2 regularly applied penalties on layer parameters or
layer activity during optimization.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
generation of AVs. Consequently, the design of self-driving
vehicles must adhere to stricter standards than the ones
existing [32-36].
The simulation results given in [37] indicate that the
proposed scheme can effectively increase the network
throughput for LTE-A small-cell networks with dual-
connectivity enhancement. In ITS, dual connectivity supports
cybersecurity solutions with intelligent verification.
According to [20], the optimization of accuracy in ITS and
intelligent AV (IAV) is the strongest measurement of finding
a decision for technical issues and developments. Here, all
calculations and measurements must be accurate, with precise
values optimized through an efficient optimization method.
According to [5], with blockchain technology, examining
enterprise security policy maximizes the strength of the
security levels, providing a quality service when hackers’
attacks affect the medical data of hospitals. Updating security
policy intelligently protects the confidential data of
organizations. In addition, managing an intelligent security
policy allows users to address the security risks associated
with the 6G generation. Building a taxonomy to enhance
automotive system security [38] supports the security issues
considered in vehicular networks. AI-based security solutions
have also been updated with intelligent security policies to
enhance automotive systems security. A congestion-aware
pre-predictive data-allocation model [39] was used to improve
the cooperative intelligent transportation system. This model
depends on the intelligence level that can be created from
predictive data management employing 6G communication
and computation methods.
According to [40], 6G-based intelligent cybersecurity will
lead to new techniques; some of these are given below.
1) Cryptographic hash drones are employed to enhance
intelligent cybersecurity solutions in AVs and
autonomous mobile systems.
2) Lightweight authentication techniques with IC6G-
based policies and AI-based emerging technology,
such as 6G-based complex networks
3) AI-based cybersecurity techniques with advanced
security protocols based on photonic sensor networks
and quantum cryptography for autonomous vehicular
communication
4) Intelligent cellular technology (7G) can enhance AI-
based cybersecurity solutions used in AV.
The requirement for 6G safety and intelligence of AVs will
improve with the development of 6G and progress as security
demands, as shown in Fig. 2.
Finally, several studies have focused on detection
performance in mobile environments [2], which is important
for enhancing cybersecurity, encrypting medical images
against various threats when transmitting data via wireless
broadcasting [41], and using deep-learning algorithms in
segmentation tasks with various kinds of networks [42].
A. AUTOMATING THE ADOPTION OF MACHINE
INTELLIGENCE WITH POLICIES
All systems work with standard operating policies which
provide insurance to all devices and autonomous systems. By
using machine-intelligent programs, policies can be
maintained according to users’ requirements. Adopting
machine intelligence with policies will increase cybersecurity
solutions since all AV user transactions must be registered. For
instance, attacks using ransomware will be difficult because
automation with machine intelligence will monitor all
transactions intelligently with policies set by the service
FIGURE 2. Future of 6G safety influenced to intelligent vehicular network [ 21].
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
providers and users. These policies should be set at least 24
hours before the actual and specific transactions. Existing
models use anomaly detection (Fig. 3) to improve the security
of AVs. However, intelligent cybersecurity solutions can also
be enhanced using anomaly detection and other rules, such as
policies. Using our proposed model, users who are vulnerable
and elderly will be supported when they use AVs through
strong intelligence policies. Adopting machine intelligence
with emerging IC6G technologies in combination with
specific policies is key to improving the future of
cybersecurity solutions. Strong policies secure the public
environment, which also includes the banking sector.
B. AN OVERVIEW OF INTELLIGENT CYBERSECURITY
Emerging technological trends have been focused on the many
flexible features of 6G based security devices used in AV
where we can add intelligent security solutions, such as
intelligent cybersecurity. Attacks on the 6G architecture and
6G-based emerging networks (Fig. 4) will affect the services
used in Avs if service providers do not employ the appropriate
or proactive security mechanisms. Therefore, 6G architecture
should be secured using a 6G-based intelligent cybersecurity
model. In this paper, IC6G is portrayed with the combined
features of intelligent and cybersecurity solutions for AVs.
This novel usage of IC6G and its emerging technologies in AV
will enhance EE and overall security performance.
As shown in Fig. 5, the following attacks illustrate the
security issues in the AVs that rely on 6G-based intelligent
services and cybersecurity solutions:
1) Adversarial attacks: All traffic signals and
communication channels between the vehicles and service
providers, such as banks, should be cleaned and secured
dynamically by the AV’s intelligent service providers.
2) Data poisoning: All transactions depend on data that
comes from many different sources but has been cleaned for
service creation. Here, an injection poisons the data and must
be removed from the communication channels of V2X and
AVs.
3) Compromised UAVs: Fake GPS information damages
all services, including communication links between users.
4) Sybil attacks: Virtual traffic jams create signal
interference between users. A significant attack on
autonomous vehicle networks known as a (Sybil attack)
occurs when an attacker maliciously assumes or steals several
identities and utilizes those identities to disrupt the AVs’
network's functionality by spreading fictitious identities. The
research model should be able to detect these attacks and
provide the best services to all AV users.
Different categories of policies and delivery times of reports
influence the policies created at each level of intelligence. The
following levels of intelligence prevent attacks, threats, and
vulnerabilities (Table III). These levels of intelligence provide
an automation system that can adjust the machine’s
intelligence, allowing it to identify vulnerabilities proactively.
TABLE III
INTELLIGENCE THAT DEPENDS ON POLICIES
Intelligence
Level
Policies
Description
1
Enacted
according to the
situation
Users’ regular time, place, cost of
the transaction, frequency of use,
etc.
2
Ensure timely
and confidential
delivery of
policy
Authorized items should be
delivered on time with the
tracking scheme (intelligent
cybersecurity through
management)
3
Based on the
operational
conditions of
the devices
Technical requirements which
affect IC6G, and the cybersecurity
solutions integrated in AVs
FIGURE 3.
Workflo w of hyb rid deep an omaly detection approach [4].
FIGURE 4. 6G landscape and security composition [43].
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
III. PROPOSED RESEARCH
The communication features, services, and transactions
integrated within autonomous vehicles should all be secured
using the proposed model. The constant evolution of
cyberattacks has been taken into account in the problem
statement; as such, the proposed model should detect such
attacks instantly. To resolve all possible security problems, the
IC6G approach with suitable security algorithms was
employed in the proposed model.
A. PROBLEM STATEMENT
AVs have one of many compulsory services involved with
money transactions for automatic charges when using
autonomous vehicles. Many users have reported the loss of
millions of dollars after paying charges for bogus services
while traveling. Hackers act as authorized persons and steal
users’ money, and unfortunately, banks are unable to directly
stop those transactions, as they still operate under the
assumption of protecting deposited money from thieves,
hackers, and physical violators. The problems this creates are
many, and the services established by the service providers
and the providers’ policies create even more cybersecurity
problems, as they inadvertently support hackers. Thus, these
policies should be handled intelligently and according to the
situation, location, time and other major relevant factors.
In cyberattacks such as phishing, solutions with a 6G-based
intelligent cybersecurity model can solve these problems
intelligently and proactively. Scientists have developed many
cybersecurity solutions for many illegal activities, but it is the
policies that block personal interests and encourage hackers to
get involved in illegal activities when they see the ease with
which these transactions can be attacked.
In autonomous vehicles, the following policies are executed
proactively when the system works intelligently (Table IV).
When these policies are handled intelligently and with
political support, each transaction can be secured. The policies
enacted should protect both users and service providers from
the vulnerabilities created by the communication devices used
in AVs. Further, these policies should encourage service
providers to make the necessary decisions proactively.
TABLE IV
EXAMPLES OF POLICIES INFLUENCED BY INTELLIGENT CYBERSECURITY
Policies
Description
Limits should be controlled
They can be controlled by the intelligent
system rather than AV users. The IC6G
will have proactive security solutions
Accessing features or
services with authorized
codes
All services must be monitored with time,
type of service, etc. Machine intelligence
will record all transactions proactively
Maximum transactions per
day
Intelligent systems should verify both the
senders’ and receivers’ details. Reliance
on IC6G to do so with updated policies
Proper security codes for
each transaction when
exceeding the limit
A receipt should be exchanged clearly
with the authentication and authorization
codes. Fake users will be blocked from
entering any intelligent systems
Intelligent sensors placed peripherally around the AV are in
direct contact with the AV’s electronic devices, including the
communicating transmitters and receivers. An intelligent
cybersecurity model detects the vulnerabilities of these
devices when they face cyberattacks and threats.
The energy consumption,
!!
, is a function of several
transceiver variables, with the most important variable being
distance,
"
, and is summarized as
!!# !"! $%&"#
(1)
In (1),
!"!
is the distance-independent term that accounts for
the overhead of radio electronics and digital processing.
%&"#
is the distance-dependent term, where
%
stands for the
amplifier inefficiency factor,
&
is the free-space path loss,
"
is the distance, and
'
is the environmental factor.
'
can be set
as a number between 2 and 4 depending on the condition of
the environment and the vulnerability of devices and
communication channels;
%
determines the inefficiency of the
transmitter when producing maximum power
&"#
at the
antenna. Energy (() is equal to the multiplication of power ())
and time (*).
!! # +!$,!%-.//0
(2)
Here,
!%
and
!$#1 !%21 !!
are the input and output
energy, and they verify the EE of the overall model with (2).
The policies and level of intelligence change, thus,
verifications depend on the vulnerabilities of the devices used
in the AVs. Intelligence and policies affect not only the
vulnerabilities of all components integrated within the AV but
also the communication channels from the AV. In this
research, we assume that the input parameters of (1), (2), and
(3) take different values according to the levels of policies and
intelligence.
The sum capacity (
!&
) is proportional to
!"!
; we can also
assume that
!"! # !&
because the energy during secure and
insecure communication is different due to many factors and
influences, as given in (3).
!&#
3 3
4',%
%)*
')* 567+
8
. $19',%,+:',% $ ;',% -
<
1
(3)
FIGURE 5. An illustration of four typical security attacks in 6G V2X [29].
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
The sum capacity (
!&
) of 6G-enabling technologies as
given in (3) influences
!"!
and is dependent on the network
coverage (k), bandwidth (B), loss noise (N), loss interference
(I), channels (i), and power (P). Intelligent cybersecurity
depends on these parameters, which work with the policies
and conditions of operations adjusted according to natural
attacks and internal and external security issues. When
accounting for all vulnerabilities, the overhead increases with
the level of intelligence, which is dependent on the operation
of services, which in turn influences the policies that service
providers set.
B. PROPOSED MODEL
In this research, an autonomous service is considered an
example of a feature integrated within the proposed model.
The proposed method used the model developed in this study,
as shown in Fig. 6. In this method, intelligent cybersecurity is
considered using intelligent features and the IC6G policies.
The proactive AV features and IC6G-based policies
considered in the proposed model were implemented in the
novel design of this method. Intelligence-based policies are
created from available or collected data related to intelligence-
dependent services. In this study, we collected data from
service users who were influenced by cyberattacks. In the 6G-
based intelligent cybersecurity solution, network security
algorithms incorporating intelligent techniques developed
from applied cryptography were used.
All cybersecurity policies that allow service providers to
secure their services will be considered in the following
section, where the results will be focused on the reflection of
those policies. According to (3), 6G-based intelligent
cybersecurity solutions (Fig. 7) depend on the policy and
conditions of the parameters used in (3).
To secure a user’s identity or personal information, a
Remote Procedure Call (RPC) can be used to secure remote
procedures with an authentication technique. The host and the
user who is requesting a service are both authenticated through
the Diffie-Hellman authentication technique. Data Encryption
Standard (DES) encryption is used by that authentication
mechanism.
Here is a scenario: Travelers can use autonomous vehicles
for short visits or other such journeys. After a long day, the
user or traveler is tired and sleeps during the journey. When
they finally arrive at home, they receive a call from a visa
office regarding identity verification of a visa they had applied
for. Tired, they take the call, not realizing that it is not genuine,
and answer “Yes” to their questions, after which they go back
to sleep. This was, in fact, a call by a hacker. The next
morning, they wake up to messages from the bank, and upon
checking their bank account, find that their money has been
stolen by the hacker. According to the messages from the
bank, 18 transactions happened during that night from that
single “Yes.” In this situation, what are the bank’s and account
holders’ responsibilities?
The bank should have contacted the client personally and
verified the situation. If their phone was switched off or if the
bank was unable to contact the person during the night, the
bank should have stopped all transactions; what happens
instead is that the blame is directed solely at the account holder
for having said “Yes”. In this situation, the
user/traveler/account holder could not have done anything
because they were unaware and asleep.
Many hackers find opportunities to attack when users or
passengers of AVs transfer or pay money from their accounts
to real senders or vendors. Intelligent and automated networks
supported by 6G-based communication technologies enhance
the cybersecurity solutions during transactions established
between the 2 authorized nodes (sender and receiver). Here,
6G-based intelligent cybersecurity solutions depend on the
following questions, which simplify the transactions within
autonomous vehicles:
What type of AI-based cybersecurity algorithms does the
proposed model use?
How many AI-based cybersecurity algorithms does your
6G-based intelligent cybersecurity model have?
How frequently do service providers (banks) update
security policies, such as transactions limits?
How long until AI-based cybersecurity algorithms can
trigger detections in each 6G-based transaction?
FIGURE 6. The proposed model for intelligent cybersecurity solutions.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
How many 6G-based intelligent algorithms require a
learning period for normal and abnormal transactions?
How does your transaction prioritize critical and high-risk
hosts that require immediate attention from the service
provider or bank?
What is the complexity reduction that the proposed model
provides for security analysts?
IV. RESULTS
The experimental setup and actual parameters for each AV
should be considered in each result. Generally, security limits
(High, Medium, and Low) should be set either by the experts
or the intelligent approach of the systems designed by the
experts. In other words, the service providers advised by these
experts must provide the necessary security solutions that
would allow us to update the IC6G approach considered in the
AVs.
In this experiment, we collected data from 100 random
users attacked by hackers from different banks. Table V lists
the structure of the data used in this experiment. However, we
have elaborated on the details of the data sizes, columns, and
rows considered in this table. Moreover, 70% of bank users
are attacked a few times (less than 3% of the users within a
fixed time) by hackers when the security limit is set to the low
bank balance of the users. In addition, 20% of bank users are
attacked several times (less than 17% of the users within a
fixed time) by hackers when the security limit is set to the
medium bank balance of the users. Finally, 10% of bank users
are attacked more times (less than 50% of the users within a
fixed time) by hackers when the security limit is set to the high
bank balance of the users. To improve the results, 6 random
places where international banks are located were chosen
when AV is moving. The average percentage of all 3 security
limits when hackers’ activities are involved is recorded in
Table V.
TABLE V
HACKERS’ ACTIVITIES AGAINST AUTONOMOUS VEHICLE USERS WHO WERE
ATTACKED
User
1
User
2
User
3
User
4
User
5
User
6
Low
(70 users)
2%
2.5%
2.4%
1.9%
1.7%
1.2%
Medium
(20 users)
15%
10%
11%
17%
9%
14%
High
(10 users)
31%
43%
49%
27%
34%
42%
The different security limits are sometimes set according to
a user’s earnings and preference and are set by the users. In
many places, it is set by the banks or systems authorized by
expert service providers. Within the current system of bank
transactions for paying expenses and services, clues were left
that indicated they were hacked. In these studies, people who
kept their withdrawal limit low never lost their money but
were still attacked in multiple ways. The people with a
medium limit had mixed attacks (2% lost the money, 15%
were attacked, but did not lose money) in public locations,
where they were most probably targeted by expert hackers
who were sacked from public organizations. People with high
limits were also attacked by hackers; in those cases, a high
limit was set by the service providers without the users’
official authorization.
Fig. 8 shows the different security limits when an AV faces
cyberattacks or threats, classified into the following
categories:
1. High limit: The threats encountered by the high limit
tend to damage the configurations of the communication
services, which include services such as transferring cash for
users’ expenses. This specific feature, integrated as AV
onboard diagnostics (OBD), sends a warning when a high
limit is set. The limits may be set by the bank or users or
autonomous system, but they must be set intelligently and
recorded with maximum evidence or verifications and/or
mutual understanding of users. These recorded verifications
must be kept at least a few weeks for minimizing illegal
transactions. When we use the IC6G approach in autonomous
FIGURE 7. Security issues and 6G-based intelligent cybersecurity solutions.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
vehicles, users get the correct information on verification
procedures through the OBD.
2. Medium limit: These threats weaken and slow down
the communication services of AVs. In all communication
services, both users and service providers should be alert
during the number of continuous transactions.
3. Low limit: Selected threats, such as cyberbullying, may
be extracted from the profiles of users because the transaction
is set to a low limit. It is the users’ responsibility.
As shown in Fig. 8 and 9, the policies of the devices used in
an AV will change the vulnerabilities and secrecy rate of the
services, respectively. Using IC6G, the overall security
facilities of an AV can be better maintained dynamically and
proactively.
All policies set for improving cybersecurity solutions need
to be reviewed according to the users’ financial circumstances.
The service providers’ responsibilities should be to support all
depositors who expect protection and security above other
facilities.
According to [44], the parameters considered for determining
vulnerabilities (Fig. 8) are proportionally equal to energy
consumption, as given in (1). The parameters given in (3) are
dependent on the policies of technical and operational limits
which affect the sum capacity (
!&
) and energy consumption
(
!"!
) of devices used in AVs.
The results of this research depend on the policies written
by experts and expert systems intelligently. The management
of financial transactions by AVs is seen as an illustration of an
intelligent cybersecurity solution based on 6G. The proposed
model's cybersecurity solutions rely on the intelligence levels
which would in turn influence policies. As shown in Fig. 9, the
results of the proposed model show 5 different services:
banking (Service 1), ticketing (Service 2), school fees (Service
3), hospital charges (Service 4), and parking payment (Service
5). In this comparison, EE is considered for the proposed
(IC6G) and 2 other (5G and 5G+ with cybersecurity (CS))
existing schemes.
Assume that all services are policy-dependent, and these
policies support the levels of intelligence considered in the
solutions of intelligent cybersecurity integrated with AVs.
Intelligence, security, complexity, energy efficiency,
trustworthiness, scalability, and privacy were used in this
study. The following explanations are provided below.
• Intelligence: Although the behavior of the same user is
acceptable, intelligence can be noted from policies or
keywords entered in the field of the service. Furthermore,
intelligence analyzed against policies or keywords
depends on the previous behaviors of users when the
service is being used.
• Security: Strong policies increase the security of all
services when cyberattacks occur during mobile
transactions. The automation of these policy generations
will improve the security of services considered in AVs
with some delays, which is the trade-off between policy
and security.
• Complexity: The complexity increases when users expect
maximum security because there is a tradeoff between the
cost of energy and security.
• Energy efficiency (EE): Analyzing the enhancement of
EE with the complexity and intelligence levels and the
strength of the policies is a common technique for
enhancing security.
• Trustworthiness: The reputation of the packet and its
trustworthiness are evaluated based on one or more of the
four verifications: data quality, location of service users,
time of accessing services, and travel direction of the AV.
• Scalability: The use of sensors with intelligent
cybersecurity increases when more service users and AVs
are involved.
• Privacy: Policies will also enhance intelligent
cybersecurity because some of the data used in
automated and connected vehicles are personal and
sensitive.
V. DISCUSSION AND ANALYSIS
Although appropriate cybersecurity solutions are assessed in
this study, the following points are noted as having a
substantial impact on the outcome results, as they provide zero
or minimum cybercrime, which can result in loss of control of
FIGURE 9. Results of the proposed model.
FIGURE 8. Vulnerabilities as 𝑬
𝒅
against different cyberattacks based on
security limits.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
critical equipment used in AVs. Furthermore, cybercrime
attacks the warning systems responsible for services
integrated into AVs. In addition, they can cause damage to
human health and the environment resulting from catastrophic
spills, waste discharges, and air emissions.
All results we obtained in this research depend on the
limits set by the intelligent experts who provide the intelligent
cybersecurity solutions to many sectors such as business.
Within the business sectors, banking system is considered as
an example or scenario in these results. Although many
sectors and systems (medical, business, etc.) use the secure
services through the intelligent cybersecurity, we have
considered some selected services in this result. Intelligent
cybersecurity solutions vary with the EE affected by the
security limits and vulnerabilities Although 5 services are
considered in the results, a specific service is to provide the
necessary discussion and analysis. In some international banks
and their services, the transferring procedures of the policies
used in the system need to be investigated, as they are the real
problem. A hacker can fool people and transfer millions of
dollars ($) or Saudi riyals (SR) within a minute if the
transferring policy in some international banks is not secured.
For example, a hacker can act as a legal officer and ask for
verification from a person who has paid visa fees from their
account to an official account. The average person trusts third
parties in many situations and circumstances to enact such
payments. Intelligent experts and systems should have some
procedures which depend on the policies, steps, and evidence
collected from banks. To design and develop the intelligent
procedure, the following evidence is collected from the bank:
• The receiver’s account details were not properly checked;
the receiver can open the account and delete the account
without references.
• The senders’ confirmation must be verified personally for
securing the transactions.
• The bank must have the proper verifications before
sending the one-time password.
• Account holders must trust the banks, but banks must not
trust the receivers without proper verifications.
• The bank should make sure that the receiver’s account
number is active for at least the last 3 months and valid
for at least the next 3 months after the transactions.
Among the many services used in AVs, communication
services are deployed for users who would like to
communicate or exchange online transactions when they pay
for their expenses during a journey. Users should be able to
use the services (banking, ticketing, schooling (Tuition and
other fees for academic services), etc.) comfortably and
securely. In this discussion, 5 different services are
considered, as previously mentioned: services 1, 2, 3, 4, and 5,
banking, ticketing, school fees, hospital charge, and parking
payment, respectively. When we deploy the IC6G approach in
our proposed model, all 5 services are improved because
policies are set up intelligently according to users’ financial
situation and transaction history. Whatever the situation, one
of all 3 security limits should be selected and issued
intelligently, instantly, and dynamically by the service
provider. If the account holder’s phone is switched off, but the
bank has allowed the hackers to transfer money (the bank
should have waited until verbal confirmation from the account
holder).
In this discussion, the evidence mentioned above should be
considered carefully to improve security when transferring or
withdrawing money from an account. In addition, we
proposed a model with solutions using the IC6G-based
policies to prevent cyberattacks and cybercrimes. The bogus
services during movement, unintelligent behaviors, and the
interruption of the handling services attacked by hackers are
the problems discussed in this study. Intelligence levels were
obtained from the policies concluded by the previous
behaviors of the users of the services. We solved the research
problem by analyzing intelligence levels with these policies.
VI. CHALLENGES AND LIMITATIONS
The AVs with an IC6G will have many challenges which
affect the users’ daily life. The architecture we proposed in this
research will present new opportunities for many potential
systems and future applications:
• Autonomous vehicles’ basic and luxury features will
influence 6G-based gadgets.
• An introduction of cybersecurity solutions in 6G
networks and related platforms used in autonomous
vehicles.
• An increase in intelligent features and proactive
cybersecurity solutions.
The above points will spur research that support improving
transportation policies.
Brain Controlled Vehicles (BCV) may be introduced for
simplifying the operations of the devices used in autonomous
systems, including AVs. Further, the functions of 6G networks
will make BCVs possible and will support IC6G in improving
the intelligent features of the AVs.
Regarding the cost of energy and intelligent cybersecurity,
the most challenging aspect of cost and EE is determining the
trade-off between five aspects:
i. Evolution of AV technology
ii. Access to AV technology by stakeholders
(communication service providers, road operators,
automakers, AV consumers, repairers, and the general
public)
iii. Limiting hackers' access to AV technology
iv. Widespread dynamic strategy for avoiding hacker
amplification
v. Efficient usage of AV operating logfiles [45].
According to [46, 47], tons of CO2 emissions and millions
of hours of driving every year will be saved with AVs, creating
vulnerabilities in the communication devices used in those
AVs. To solve these challenges, we need tough security
policies that need to be applied intelligently. Our research
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
model and approach provide a basic idea: intelligent
cybersecurity with machine learning and AI algorithms should
be considered to solve these problems. Intelligent
cybersecurity with UAVs may offer some unique security
challenges to 6G networks, especially regarding AVs used on
land; it is possible that advances in UAV will lead to AVs
getting low-cost energy and security.
The strength of this work lies in the IC6G policies, which
should be the best for improving cybersecurity solutions
because these policies are generated from the users’ behaviors
noted in each previous handling of the services. For instance,
the limits and changes in bank transactions in banking services
are noted to generate policies.
On the other hand, the weakness and limitations of the
research lie in the collection of previous behaviors for the last
7 days to 3 months, which will increase the time complexity
and storage, creating unnecessary delays when services are
being used during the transactions. In addition, there are
several other limitations regarding the cost of energy and
intelligent cybersecurity for users and others: the collection of
confidential data and generated policies depends on the
behavior of the previous history of the services allocated in the
Avs and regarding the importance of licensed details of the
final official 6G release in relation to the IC6G-based policies.
VII. CONCLUSIONS AND FUTURE WORK
This study presented the results from the proposed model
that might be effective for intelligently detecting and thwarting
cyberattacks on AVs and intelligent cybersecurity solutions
that maintain secure services from all vulnerabilities created
by attackers, faulty devices, or fake messages.
Policies developed for AVs should enhance the protection
of all users and communication devices integrated within the
Avs. When securing service policies are maintained by
intelligent experts, both users and service providers can
secure services using a proactive approach. As the strength
of the policies increases, the intelligence level also provides
more intelligent cybersecurity solutions.
Therefore, the security limits discussed in the results
should be set and fit by service providers based on the
situation and important security factors, such as
authentication.
The main contribution of the proposed approach is
intelligent cybersecurity solutions that provide the necessary
security to all services used in AVs when cyberattacks occur.
Furthermore, cyberattacks affect the electronic functions of
AVs, which damage the AVs’ operations and maneuvering of
vehicle movements. The influence of intelligent
cybersecurity not only solves the AVs safety issues of
electronic control systems, but also provides secure services
to passengers using the AV.
Insights from this study are provided through the proposed
model, which includes 6G-based cybersecurity solutions and
policies. Intelligent cybersecurity is considered to maximize
security and minimize energy costs for all passengers using
autonomous and mobile services while traveling. The
proposed solutions use IC6G-based policies to prevent
cyberattacks and cybercrimes and intelligently enhance the
effectiveness of cybersecurity solutions.
In this paper, previous researchers and authors provided an
overview of IC6G and the related emerging technology in
autonomous vehicles, proposed a taxonomy for IC6G through
a thorough literature review, presented a conceptual model for
IC6G to improve the level of security solutions in AVs with
cutting-edge integrated devices and technology, and presented
the challenges and issues for the discussion of novel IC6G
applications.
Furthering the work of the proposed model, we can add
more features and services to keep up with the emerging
security technology as long as it is suitable for the situation
and environmental conditions. Securing future services with
intelligent cybersecurity in AVs will depend on emerging
security technology (7G) and the strength of policies at the
time. Furthermore, these features and services depend on
energy-efficient algorithms and emerging technologies
considered at the time. This research will continue to develop
AVs with intelligent vision and ‘human-like’ thinking
capabilities.
ACKNOWLEDGMENT
This project was funded by the Deanship of Scientific
Research (DSR), King Abdulaziz University, Jeddah, under
grant no. G-601-611-39. The authors, therefore, gratefully
acknowledge with thanks the DSR for technical and financial
support.
REFERENCES
[1] Brovarone, Elisabetta Vitale, Jacopo Scudellari, and Luca
Staricco. "Planning the transition to autonomous driving:
a policy pathway towards urban liveability." Cities 108
(2021): 102996.
[2] Wu, Yirui, Haifeng Guo, Chinmay Chakraborty,
Mohammad Khosravi, Stefano Berretti, and Shaohua
Wan. "Edge computing driven low-light image dynamic
enhancement for object detection." IEEE Transactions on
Network Science and Engineering (2022).
[3] Khan, Shah Khalid, Nirajan Shiwakoti, Peter
Stasinopoulos, and Matthew Warren. "Dynamic
assessment of regulation and policy framework in the
cybersecurity of Connected and Autonomous Vehicles."
Australasian Transport Research Forum ATRF 2021-
Proceedings, 2021
[4] Prathiba, S.B., Raja, G., Anbalagan, S., Arikumar, K.S.,
Gurumoorthy, S. and Dev, K., 2022. A Hybrid Deep
Sensor Anomaly Detection for Autonomous Vehicles in
6G-V2X Environment. IEEE Transactions on Network
Science and Engineering.
[5] Nafchi, M.A. and Shahraki, Z.A., 2022. IT Governance
and Enterprise Security Policy in the 6G Era. In Next-
Generation Enterprise Security and Governance (pp. 227-
245). CRC Press.
[6] Ramezanpour, K. and Jagannath, J., 2022. Intelligent zero
trust architecture for 5G/6G networks: Principles,
challenges, and the role of machine learning in the
context of O-RAN. Computer Networks, p.109358.
[7] She, Changyang Chengjian Sun, Zhouyou Gu, Yonghui
Li, Chenyang Yang, H Vincent Poor, and Branka
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
Vucetic. "A Tutorial on Ultrareliable and low latency
communications in 6 G integrating domain knowledge
into deep learning Proceedings of the IEEE 109 no 3 2021
204 246.
[8] Khan, Shah Khalid, Nirajan Shiwakoti, Peter
Stasinopoulos, and Matthew Warren. "Security
assessment in Vehicle-to-Everything communications
with the integration of 5G and 6G networks." In 2021
International Symposium on Computer Science and
Intelligent Controls (ISCSIC), pp. 154-158. IEEE, 2021.
[9] Basnet, Manoj, and Mohd Ali. "A Deep Learning
Perspective on Connected Automated Vehicle (CAV)
Cybersecurity and Threat Intelligence." arXiv preprint
arXiv:2109.10763 (2021).
[10] Shrestha, Rakesh, Atefeh Omidkar, Sajjad Ahmadi Roudi,
Robert Abbas, and Shiho Kim. "Machine-learning-
enabled intrusion detection system for cellular connected
UAV networks." Electronics 10, no. 13 (2021): 1549.
[11] Jiang, Wenbo, Hongwei Li, Sen Liu, Xizhao Luo, and
Rongxing Lu. "Poisoning and evasion attacks against
deep learning algorithms in autonomous vehicles." IEEE
transactions on vehicular technology 69, no. 4 (2020):
4439-4449.
[12] Shrestha, Rakesh, Rojeena Bajracharya, and Shiho Kim.
"6G enabled unmanned aerial vehicle traffic
management: a perspective." IEEE Access 9 (2021):
91119-91136.
[13] Khan, Latif U., Yan Kyaw Tun, Madyan Alsenwi,
Muhammad Imran, Zhu Han, and Choong Seon Hong. "A
dispersed federated learning framework for 6G-enabled
autonomous driving cars." arXiv preprint
arXiv:2105.09641 (2021).
[14] Gui, Guan, Miao Liu, Fengxiao Tang, Nei Kato, and
Fumiyuki Adachi. "6G: Opening new horizons for
integration of comfort, security, and intelligence." IEEE
Wireless Communications 27, no. 5 (2020): 126-132.
[15] Siriwardhana, Yushan, Pawani Porambage, Madhusanka
Liyanage, and Mika Ylianttila. "AI and 6G security:
Opportunities and challenges." In 2021 Joint European
Conference on Networks and Communications & 6G
Summit (EuCNC/6G Summit), pp. 616-621. IEEE, 2021.
[16] Ghorbani, Hamidreza, M. Saeed Mohammadzadeh, and
M. Hossein Ahmadzadegan. "Modeling for malicious
traffic detection in 6G next generation networks." In 2020
International Conference on Technology and
Entrepreneurship-Virtual (ICTE-V), pp. 1-6. IEEE, 2020.
[17] Haseeb, K., Rehman, A., Saba, T., Bahaj, S.A., Wang, H.
and Song, H., 2022. Efficient and trusted autonomous
vehicle routing protocol for 6G networks with
computational intelligence. ISA transactions.
[18] Khan, S.K., Shiwakoti, N. and Stasinopoulos, P., 2022. A
conceptual system dynamics model for cybersecurity
assessment of connected and autonomous vehicles.
Accident Analysis & Prevention, 165, p.106515.
[19] Basnet, M., Poudyal, S., Ali, M.H. and Dasgupta, D.,
2021, September. Ransomware detection using deep
learning in the SCADA system of electric vehicle
charging station. In 2021 IEEE PES Innovative Smart
Grid Technologies Conference-Latin America (ISGT
Latin America) (pp. 1-5). IEEE.
[20] Gao, Yongbin, Fangzheng Tian, Jun Li, Zhijun Fang, Saba
Al-Rubaye, Wei Song, and Yier Yan. "Joint Optimization
of Depth and Ego-Motion for Intelligent Autonomous
Vehicles." IEEE Transactions on Intelligent
Transportation Systems (2022).
[21] Nguyen, Van-Linh, Ren-Hung Hwang, Po-Ching Lin,
Abhishek Vyas, and Van-Tao Nguyen. "Towards the Age
of Intelligent Vehicular Networks for Connected and
AVin 6G." IEEE Network (2022).
[22] Renda, A., Ducange, P., Marcelloni, F., Sabella, D.,
Filippou, M.C., Nardini, G., Stea, G., Virdis, A., Micheli,
D., Rapone, D. and Baltar, L.G., 2022. Federated
Learning of Explainable AI Models in 6G Systems:
Towards Secure and Automated Vehicle Networking.
Information, 13(8), p.395.
[23] Tang, Fengxiao, Yuichi Kawamoto, Nei Kato, and Jiajia
Liu. "Future intelligent and secure vehicular network
toward 6G: Machine-learning approaches." Proceedings
of the IEEE 108, no. 2 (2019): 292-307.
[24] Porambage, Pawani, Gürkan Gür, Diana Pamela Moya
Osorio, Madhusanka Liyanage, Andrei Gurtov, and Mika
Ylianttila. "The roadmap to 6G security and privacy."
IEEE Open Journal of the Communications Society 2
(2021): 1094-1122.
[25] Kim, Kyounggon, Jun Seok Kim, Seonghoon Jeong, Jo-
Hee Park, and Huy Kang Kim. "Cybersecurity for
autonomous vehicles: Review of attacks and defense."
Computers & Security (2021): 102150.
[26] Sun, Xiaoqiang, F. Richard Yu, and Peng Zhang. "A
Survey on Cyber-Security of Connected and
AV(CAVs)." IEEE Transactions on Intelligent
Transportation Systems (2021).
[27] Gao, Cong, Geng Wang, Weisong Shi, Zhongmin Wang,
and Yanping Chen. "Autonomous Driving Security: State
of the Art and Challenges." IEEE Internet of Things
Journal (2021).
[28] Chow, Man Chun, Maode Ma, and Zhijin Pan. "Attack
Models and Countermeasures for Autonomous
Vehicles." In Intelligent Technologies for Internet of
Vehicles, pp. 375-401. Springer, Cham, 2021.
[29] Campisi, Tiziana, Alessandro Severino, Muhammad
Ahmad Al-Rashid, and Giovanni Pau. "The development
of the smart cities in the connected and AV(CAVs) era:
From mobility patterns to scaling in cities."
Infrastructures 6, no. 7 (2021): 100.
[30] Ribeiro, Manuel Alector, Dogan Gursoy, and Oscar
Hengxuan Chi. "Customer Acceptance of AVin Travel
and Tourism." Journal of Travel Research (2021):
0047287521993578.
[31] Aldakkhelallah, Abdulaziz, and Milan Simic. "AVin
Intelligent Transportation Systems." In International
Conference on Human-Centered Intelligent Systems, pp.
185-198. Springer, Singapore, 2021.
[32] Vijey Thayananthan, “Advanced security issues of IoT
based 5G plus wireless communication for Industry 4.0”.
https://novapublishers.com/shop/advanced-security-
issues-of-iot-based-5g-plus-wireless-communication-
for-industry-4-0/ (ISBN: 978-1-53615-538-9)
[33] Shaikh, Riaz Ahmed, and Vijey Thayananthan, Patent:
Trust evaluation wireless network for routing data
packets (US10225708B2) granted on 5th March 2019.
https://patents.google.com/patent/US10225708B2.
[34] Algarin, Abdullah, and Vijey Thayananthan.
"Improvement of 5G Transportation Services with SDN-
Based Security Solutions and beyond 5G." Electronics
10, no. 20 (2021): 2490. ISI impact factor 2.412
[35] Shaikh, Riaz Ahmed, and Vijey Thayananthan. "Risk-
Based Decision Methods for Vehicular Networks."
Electronics 8, no. 6 (2019): 627.
[36] Thayananthan, Vijey and Javad Yazdani. “Secure Cyber-
Physical Systems for improving transportation facilities
in Smart cities and industry 4.0.” (ISBN13:
9781522571896).
[37] Pan, M.S.; Lin, T.M.; Chiu, C.Y.; Wang, C.Y. Downlink
Traffic Scheduling for LTE-A Small Cell Networks with
Dual Connectivity Enhancement. IEEE Commun. Lett.
2016, 20, 796–799.
[38] Haddaji, A., Ayed, S. and Fourati, L.C., 2022. Artificial
Intelligence techniques to mitigate cyber-attacks within
vehicular networks: Survey. Computers and Electrical
Engineering, 104, p.108460.
[39] Manogaran, G., Alrayes, I., Alshaikhi, A. and Rawat,
D.B., 2022. Pre-Predictive Congestion-Based Data
Allocation for Sixth Generation Cooperative Intelligent
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME XX, 2017 9
Transportation Systems. IEEE Transactions on Intelligent
Transportation Systems, 23(10), pp.18655-18667.
[40] Khan, A.S., Sattar, M.A., Nisar, K., Ibrahim, A.A.A.,
Annuar, N.B., Abdullah, J.B. and Karim Memon, S.,
2023. A Survey on 6G Enabled Light Weight
Authentication Protocol for UAVs, Security, Open
Research Issues and Future Directions. Applied Sciences,
13(1), p.277.
[41] Wu, Yirui, Lilai Zhang, Stefano Berretti, and Shaohua
Wan. "Medical image encryption by content-aware DNA
computing for secure healthcare." IEEE Transactions on
Industrial Informatics (2022).
[42] Shi, Guangchen, Yirui Wu, Jun Liu, Shaohua Wan,
Wenhai Wang, and Tong Lu. "Incremental few-shot
semantic segmentation via embedding adaptive-update
and hyper-class representation." In Proceedings of the
30th ACM International Conference on Multimedia, pp.
5547-5556. 2022.
[43] Porambage, Pawani, Gürkan Gür, Diana Pamela Moya
Osorio, Madhusanka Livanage, and Mika Ylianttila. "6G
security challenges and potential solutions." In 2021 Joint
European Conference on Networks and Communications
& 6G Summit (EuCNC/6G Summit), pp. 622-627. IEEE,
2021.
[44] Mir, Nader F., Computer and communication networks
Second edition. Pearson Education, Inc. publishing as
Prentice Hall (2015), ISBN 978-0-13-381474-3.
[45] Khan, S.K., Shiwakoti, N., Stasinopoulos, P., and
Matthew, W., 'Governing Connected and Automated
Vehicles: Cybersecurity Regulations and Operational
Framework ', 2022.
[46] Osorio, Diana Pamela Moya, Ijaz Ahmad, José David
Vega Sánchez, Andrei Gurtov, Johan Scholliers, Matti
Kutila, and Pawani Porambage. "Towards 6G-Enabled
Internet of Vehicles: Security and Privacy." IEEE Open
Journal of the Communications Society 3 (2022): 82-105.
[47] Obaid, M.S., 2022. Macroscopic modelling of the effects
of autonomous vehicles and cooperative intelligent
transport systems.
ABDULLAH M. ALGARNI received the Ph.D.
degree in Computer Science from the College of
Natural Sciences, Colorado State University,
USA, in 2016, and his master’s degree in
Computer Science from Colorado State University
in 2014, and another master’s degree in Software
Systems Engineering from the University of
Melbourne, Australia, in 2008. He is currently an
Associate Professor in the Computer Science
Department, King Abdulaziz University, Jeddah,
Saudi Arabia. His research interests include software engineering, software
security, and cybersecurity.
VIJEY THAYANANTHAN is a Chartered
Engineer (CEng) and he is a professor at
Computer Science Department at the King
Abdulaziz University, Jeddah, Saudi Arabia. He
obtained his Ph.D. degree in Engineering,
Communication Systems from the University of
Lancaster, UK, in 1998. Also, he worked in the
Department of Electrical and Electronic
Engineering, Glasgow/Strathclyde University,
UK as Postdoctoral Research Fellow. Since 2000,
he had been working as a research engineer and senior algorithm
development engineer in Advantech Ltd, Southampton University Science
Park, UK, and Amfax Ltd, UK respectively. His research interest includes
wireless networks, cybersecurity, information theory and coding. He is a
reviewer for various international journals, e.g., Journal of Fundamentals of
Renewable Energy and Applications, Mobile Networks and Applications
and Springer book chapters. He is a member of the IET.
This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/ACCESS.2023.3244883
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/