Cyber security Trends in Modern Automobile industry/sector

Chapter 1: Introduction

Back ground of the study

In computing, cybersecurity refers to a collection of technologies, procedures, and methods that

are intended to secure a computer system against attack or unauthorised access by hackers.
Cybersecurity has been a subject of significant importance during recent years. According to
Schmittner and Macher's (2019), a large number of autonomous cars are linked. According to the
findings of the study, these easily accessible cars are equipped with connection features that
make it easier to share information. The information exchange is primarily utilised in vehicle
maintenance as well as allowing traffic safety measures. Schmittner and Macher (2019) in his
research show that the increased network connection in autonomous cars increases the
possibilities of an attack on vehicles. They continue to explain that hackers may target the linked
infrastructure for various reasons. This, then, poses new concerns for vehicle cybersecurity.

With each passing day, the globe becomes increasingly connected. Our everyday gadgets are
now "smart" devices that are linked to the Internet and can be operated remotely from anywhere
on the planet. Smart houses, smart grids, smart equipment (e.g. smart freezers, smart washers and
dryers, smart air-conditioning) are only the beginning and even the cars we use are becoming
“smart”. Most of the modern vehicles are connected to internet in order to offer a more varied
service to the end (Eiza, and Ni, 2017). Vehicles now contain more than 100 million
programming language which is significantly bigger than an F-35 combat aircraft or a Boeing
787 passenger airliner. Recent history was made when the Tesla business introduced their latest
Model 3 and got and over 325000 bookings in a single week. This was an unprecedented
occurrence. This model includes extremely advanced software, an auto-pilot option and
numerous connecting points towards the Internet. The automotive industry will undergo
significant upheaval as the number of electric and plug-in hybrid vehicles produced rises. The
cyber-security of such automobiles is a critical problem that will only become more significant
over time (Acharya et al, 2020).

In a short period of time, vehicles have gone from being basic modes of transportation to
powerful computers on wheels that are becoming increasingly smart as well as deeply connected
into our digital lives. Automakers and customers will gain from cars becoming smarter and more
connected, but this will also expose them to greater cybersecurity risks. Because human lives are
on the line, automakers are taking cybersecurity extremely seriously and investing significant
resources into it. To identify and prevent cybersecurity threats, connected vehicles will need to
be constantly monitored. A strong security architecture that reduces attack routes and attack
surfaces is the first step in protecting cars against cyberattacks. Cybersecurity prevention will
now focus on ongoing monitoring, vulnerabilities identification, including intrusion detection
(Shan et al, 2020).

After a threat has been detected, remote over-the-air software upgrades may be utilised to
address security problems and restore protection for affected vehicle components or systems.
Autonomous vehicles (AV) as well as connected vehicles (CV) are the two kinds of intelligent
cars presently available in the autonomous sector (CV). In order to allow bi-directional
communication between connected cars, mobile devices, and other equipment, connected
vehicles join a network. Self-driving automobiles, on the other hand, are autonomous vehicles
that don't need human intervention to function. Even while almost all self-driving cars are linked
in some way, not all connected cars are automated (Amin, and tariq, 2015). Future car
technology may turn them into IoT gadgets, eliminating the need for humans to control or be in
them. Many businesses, including giants like Uber as well as Amazon, are working very hard to
make it feasible for the ordinary person to take use of this new technology. Even if the idea is
interesting, security concerns must be addressed head-on, particularly in this field. As the
number of connected cars on the road increases, safety and security experts are paying more
attention than ever. More than 18 million new autonomous cars are expected to reach the road by
2030, according to industry estimates (Frost and Sullivan). There will be increased risks for
individuals and businesses alike if automakers and suppliers don't make a concerted effort to
include security into cars from the start (Sanguino et al, 2020).

Many successful cyber-attacks against connected cars have already occurred, including violent
attacks on the Jeep Cherokee and Tesla S models as well as Nissan electric vehicles and
Chevrolet Corvettes. Because it was the first time a car has been attacked remotely, the Jeep
Cherokee assault drew the most attention. Thus, it served as a warning to the whole automobile
sector to focus more on safeguarding their cars from cyber-threats. Before this attack, the
automobile industry's primary justification for not prioritising car cyber-security was that an
attack would require accessibility to the vehicle. So the remote assault on Jeep Cherokee had a
significant effect on the automobile industry's cyber-security strategy and practices (Obaid et al,
2020).

The increasing complexity of automotive electronic systems is a direct result of the market's
increasing demands. While safeguarding its passengers is important, the market also wants the
vehicle to be safe through avoiding accidents. This is made possible with the use of driver
assistance technologies. As well as being more comfortable, the vehicle is anticipated to include
a full infotainment system and have features like automatic conditioning, seat adjustability with
memory, and automated tailgate opening (Jadhav et al, 2021). More value-added services
needing network connectivity are being offered by automakers. Automobiles in the modern day
must provide their owners a sense of continuity throughout their lives. People desire to be
constantly linked to their professional and social networks in order to access and engage with a
wide range of information. Automobile electronic designs are becoming more sophisticated and
big, especially when compared to other sectors. As a result of these needs. In comparison to
Windows 7 (39.5 million lines in 2009) or a Boeing 787, current high-end vehicle software has
100 million programming language on average (13.8 million). With that many lines of code, it's
safe to assume that some security flaws exist.

The purpose of this dissertation is to get a better understanding of the connected car's software
architecture's security measures. The study focus on AUTOSAR (Automotive Open System
Architecture), a system standard utilised by both automakers and electronic equipment vendors.
With this standard, automotive electronics architectures will be simplified, enabling businesses
to collaborate on developing the standard while competing on implementations. Embedded
controllers (ECUs) utilise AUTOSAR to provide a standardised operating system that frees
software writers from having to rely on the hardware vendor. As a de facto industry standard, it's
critical to look at it carefully from a security standpoint. The internal automotive network is a
critical component of automobiles. In the vehicle, this network links all of the electronic control
units, and each ECU manages a unique function. Because AUTOSAR doesn't need
understanding the specifics of the ECU, businesses can build software much more quickly when
using it (Ivanova, 2017).
As the automobile industry becomes increasingly aware of the dangers posed by online attacks,
so does study into the security weaknesses of this standard. The issue is that new cars for the
next 2-3 years are virtually "done," thus any study that identifies potential vulnerabilities now
may not be incorporated in vehicles until 2020. While certain software may be updated through
service centres, all programs belong to the AUTOSAR standard, and the vehicle's operating
system is built in accordance with the AUTOSAR specification. However, even if we are
successful in updating the software, each program that was previously operating on the old os
would have to be upgraded in order to function properly on the newer version of the software.
There's also the issue of data stored in the cars that's required to make them work properly. Even
if you want to keep all your apps and data operating smoothly, upgrading from Windows 7 to
Windows 10 may be difficult. However, doing it in a car that wasn't even intended to offer this
as an option is virtually impossible (Villasenor, 2011).

As a result, car software updates are still discouraged since the computer sector is much more
sophisticated than the automobile business in this regard. Because this requirement requires a
thorough cyber-security assessment, this thesis provides a threat model for it. It is the goal of the
threat modelling method to assess the security of the AUTOSAR standard and provide
recommendations on how to improve security. Current threat modelling techniques are modified
and altered in order to apply to the connected vehicle and the AUTOSAR standards before
beginning this procedure. Future versions of AUTOSAR may make advantage of the developed
threat modelling methodology in order to maintain a high degree of security in the architecture.
Each and every one of the security vulnerabilities has been recorded and classified. This data will
be useful in determining which elements of the AUTOSAR framework are weak points that need
special attention, and also protecting the program that use AUTOSAR as a foundation for their
functionality (McCarthy et al, 2014).

Problem statement

The number of linked automobiles will increase rapidly in a market that is expanding at a rate of
more about 70 million vehicles per year. As a result, hackers will have a larger attack surface to
work with. Vulnerabilities exist in software for a variety of causes. It seems that big automobile
project managers are not always well-versed on cybersecurity's significance and nuances.
Software engineers in the automobile industry aren't accustomed to considering security from the
outset of the project, according to one theory. Because they feel undertrained, they also believe
their procedures lack adequate security enabling technology. As an industry faces new
cybersecurity risks, raising security awareness and actively managing security policies may be a
problem.

Some best practises have been identified to manage the resultant risks in light of these
cybersecurity vulnerabilities, difficulties, and threats. Informed by our expertise assisting large
manufacturers to integrate cybersecurity throughout their products and operations across many
sectors, it provides sound advice. A global security plan must be developed to address these
issues. Fortunately, established technology, techniques and lessons learned and procedures in
other sectors and marketplaces exist and may be adopted and utilised in the automobile business.
Cybersecurity is a new subject. Car lifecycles are far longer than conventional computer product
lifecycles, therefore an industry-wide set of best practises is clearly needed to guarantee a safe
design over the long term. These security best practises will make the final product even more
secure.

Therefore, the study will fill the gap by analyzing the cyber security trends in automobile
industry. It will discuss in detail about different cyberattacks, standards to prevent attacks, along
with latest model of automobiles that are on highly risk.

Research aim and objective

The aim of the study is to analyse the cyber security trends in automobile industry. Moreover, the
objective of this thesis is to adapt and then apply several threat modelling methods to the
connected vehicle and the underlying AUTOSAR standard, as well as to assess their efficacy.
Researcher selected two approaches from the original study and modified them for use in the
automobile sector. Because of this, the thesis offers two threat modelling approaches that may be
used to a variety of automotive systems in the future. The findings of this thesis, as well as the
outcomes of the threat modelling method, should help to enhance cyber-security for cars that are
linked to the internet.

Research Question

 What are the current cyber security trends in automobile industry?

 What standards can be followed by the car makers to avoid threats?
  What are different type of connections that automobile car use?

Research Significance

The research is having a lot of significance because cyber security risks are at peak when it
comes to automobile sector. The findings of this study will be helpful for the car manufacturer to
design the cars by following some standards and keeping the cyber threats in mind. Therefore,
the research would be highly beneficial for major car manufacturers.
 Chapter 2: Literature Review

Cyber Security

Cybersecurity is primarily concerned with computer systems, members of the general public
seldom connect it with the automobile sector. The linked vehicle, on the other hand, hasn't been
restricted to science fiction for long and is now available from 15 different manufacturers
(Francisco, 2014). Today, many individuals can benefit from in-car entertainment systems. Wi-
Fi and Bluetooth-enabled gadgets allow drivers to take use of extra features and customise their
surroundings according on their preferences (Griffor, 2016). The automobile industry's use of
cyber security extends to linked parking and aftermarket services, among other things. As a
result, privacy and security must be ensured. Recently, hacking organisations were found to have
targeted cellphones, affecting the linked vehicle (Francisco, 2014). These concerns drew the
attention of experts, who reevaluated the severity of cybersecurity risks as a result.

The automobile industry is entering a new area of security with the introduction of self-driving
cars. Traditional computer protection has already shown itself to be difficult, and the results have
been catastrophic on many occasions. Autonomous vehicle cybersecurity is a new step that
increases system complexity and opens the door to even more catastrophic failures. However, a
small-scale cyber-attack on an autonomous vehicle has the potential to affect human lives, unlike
cyberattacks on computer systems, which usually result in financial losses. Although automotive
security is frequently considered new, it has been researched for a long time. Vehicles have
traditionally been considered to be closed systems, with electrical or digital controls being kept
separate from the mechanical systems that govern the essential components (Sanguino et al,
2020). As a result of the physical prerequisites for an assault, hackers have been unable to enter
or operate a vehicle. According to some academics and OEMs, hacks aren't possible or
susceptible enough if they need a physical link. Encryption is one kind of automobile
cybersecurity solution. Due to computational as well as data overheads, these methods have been
shown to be ineffective for the CAN protocol. Researchers have suggested alternative techniques
like Intrusion Detection Systems to overcome the shortcomings of crypto-based solutions. For
example, Hafeez et al. offer a physical-layer intrusion detection system. In order to do this, one
must first estimate the frequency range of each transmitter and thereafter train a neural network
to utilise the frequency range as a signature in order to determine the controller or transmitter.
Cyber security issues in autonomous vehicle

An AV cyberattack may be launched in a variety of methods. An attack may be directed against

the software that controls visual data and road infrastructure, or at the vehicle's hardware. A
cyberattack on remote keyless entry may lock someone inside or prevent locking (Checkoway et
al., 2011). If a hacker has control of tire-pressure monitoring devices, the systems may display
erroneous data and conceal normal air pressure leakage decrease. Because of the sensor's ability
to detect a steep incline, an assault against it may delay or stop the vehicle (Parkinson et al.,
2017).

The communication system of a car may be vulnerable. Attacks may be active, such as sending
misleading messages or interrupting communication, or they can be passive, such as collecting
information over time for use in the future. There are many current threats to automobile
communication systems. If the attacker transmits fake data or utilises a false identity, it's referred
to as a spoofing attack (e.g., they can pretend that they are a neighbouring car or send false
data about the neighbouring car location). When an attacker uses a man-in-the-middle attack,
they intercept the communication intended for the vehicle, make modifications, and then resend
it (He et al., 2017). A denial-of-service attack occurs when an intruder transmits a broad range of
information to the vehicle in an attempt to disrupt communication (Bergin, 2015). Background
radio noise is used to create a "jamming" effect, blocking a transmission frequency. An attack
known as a black hole occurs when a communication is intercepted but the vehicle is not
informed of the intercepted information (Bergin, 2015). The sender's digital signature may be
falsified, and the vehicle can be made to restart, or the car communications certificate can be
replaced with a fake ones (Petit and Shladover, 2014).

Other kinds of AV attacks rely heavily on the human element. If the car's information system is
compromised by malware, it may lead to more problems (Takahashi, 2018). The most likely
origin of the infection is an operator error. Malware (such as trojans or viruses) may initially
infect weaker systems before moving on to more critical ones (Axelrod, 2017). Putting an
infected CD in the CD player of a car may potentially lead to an attack since the malware on the
CD would be immediately downloaded (Checkoway et al., 2011). The assault that results may
cause the car's operating system to crash. Car sharing businesses often require customers to use
their cellphones to access the vehicle, thus an attacker may get access to the vehicle by hacking
the cellphone or the connection between phone and the vehicle. While malware is deployed via
social engineering or car-sharing, assaults may occur when no one is in the vehicle, making the
vehicle vulnerable to theft.

AV protection may be improved in a variety of ways, according to cyber security experts.

Countries should tighten their grip on AV-producing businesses, standardise AV technology, and
institute cybersecurity ranking measures (Lim and Taeihagh, 2018). Products sourced from
suppliers should be within the control of companies (Parkinson et al., 2017). Every element will
have its own firewall and be independent of the others in terms of communication kinds
(Messnarz et al., 2017) as well as intrusion detection system layers (Straub et al., 2017). Regular
updates of the security system are required, and CAN should be shielded from monitoring (Lim
et al., 2017). To detect a cyberattack, there may be chips added that regulate the behaviour and
temperature of various hardware components (Lima et al., 2016). A different user interface may
also increase acceptance of cybersecurity-enhancing choices. All of these mitigating strategies
should be implemented at the same time, if at all feasible.

Cyber Security and Automobile

Cybersecurity is a collection of technology, procedures, and activities aimed at preventing

hackers from gaining access to a computer system without authorization. In recent years,
cybersecurity has grown in importance. Numerous self-driving cars are linked, according to
Schmittner and Macher's investigation in 2019. Researchers found connectivity features that
make it easier to share information in these widely accessible cars. Vehicle maintenance and
traffic safety measures are made possible thanks to the exchange of information. Automated cars'
strong network connectivity, according to Schmittner and Macher (2019), raises the risk of an
attack. They go on to explain that the linked infrastructure may be attacked by hackers for a
variety of reasons. Because of this, new threats to automobile cybersecurity now exist.
According to the findings, electronic components account for almost half of the entire cost of
producing a vehicle with 100 million program code.

According to Schmittner and Macher (2019), the automobile sector has seen an annual growth of
about 30% in software-implemented functions. About 80% of new products are based on these
platforms. According to the findings, there are now about 112 million linked cars on the road
globally. As a result, the cars are more vulnerable to cyber-attacks. According to the findings of
the research, the industry for automotive cybersecurity is expected to expand at an exponential
rate over the next three years, reaching USD 759 million in revenue. The automotive sector,
according to Schmittner and Macher (2019), has chosen to invest in the development of
standards that will address the problems connected to vehicle cybersecurity and provide
protection for their assets. A proposal standard for "ISO-SAE AWI 21434 Road Vehicles -
Cybersecurity Engineering" was recently developed by the ISO and SAE standardisation bodies.
According to the findings, although this standard focuses on automobile security engineering,
other standards as well as recommendations have been released in the past. It is intended that
ISO-SAE 21434 provides a framework for engineering given the current work. It's also a
prognostication for automotive cybersecurity engineering's future development.

The researchers Miller and Valasek (2015) performed another investigation on vehicle

cybersecurity. At the outset, they claim that research aimed at improving automobile security has
gotten a lot of attention since most people own vehicles and are aware of the threats presented by
vehicle-borne assaults. In 2010, researchers from UC San Diego and UC Washington
demonstrated that they could inject messages into a vehicle's Controller Area Network (CAN)
bus. As a result, they were able to make modifications to the vehicle. The engine could be shut
off, the speedometer display could be controlled, and the brake system could be altered.

Despite the fact that Miller and Valasek (2015) found the study to be very useful, the general
public was highly critical of it. According to them, an attacker would have to get physical access
to the vehicle in order to carry out these actions. The next year, the same team of academics
collaborated on the same study and produced identical findings. They discovered that by
executing a few lines of code, they could interfere with the cars' operating system. Since so many
cars were at risk, FCA had to recall 1.4 million of them. Finally, according to Miller and Valasek
(2015), a remote assault on every car in the United States is possible. The attacker does not have
to modify the vehicle in any way, nor do they need to get physical access to the vehicle in order
for this to happen. They claim that some physical systems, like brakes and steering, are
vulnerable to the remote attack. They conclude by urging future car manufacturers to implement
more robust security measures. In order to keep drivers secure from cyber-attacks, it's the only
option. The importance of cyber security cannot be overstated in any business (Pentland et al.,
2018).
According to the studied literature, regulations governing automobile cybersecurity are
necessary. Because of the strong consumer demand for automotive technology, there is a
constant need to further computerization (Enev et al., 2016). People know that almost every new
vehicle has wireless technology. Inadequate privacy regulations and intrusion put these
technology at risk (Markey, 2015). Connected cars, as well as the gadgets they link to, hold
private information. The use of technologies in a car should go through extensive testing before
it is put into production. Policy and standards regulating automotive cybersecurity are required
because technological integration with critical automobile systems requires the formulation of
such (Schmittner et al., 2018). A set of rules supporting the worldwide automotive industry is
required for all stakeholders engaged in the distribution chain of vehicle manufacturing.

Efforts to manage automotive Cybersecurity

According to another research performed by Schoitsch et al., (2015), the advent of network
connections in automobile brought in new difficulties in terms of safety and security. The
research continues by stating that operational standards have been developed since 2000 based
upon this ISO/IEC 61508 standard. According to the research, first-generation guidelines fell
short in addressing the hazards of a highly interconnected environment. At that time, the
resulting security issues were not taken into account. A new topic to address in the study of
systems development and safety is security in a drive system, according to Schoitsch et al
(2015). The article claims that IEEE 1609.2 must be in charge of defining techniques for
securing wireless communication communications. IEEE 1609.2 This would need a change to
the IEEE 802.11 wireless automotive communication standards. According to the results of an
investigation, existing regulations for linked vehicle safety and security have numerous holes.
The ISO 26262 standards was used as an example in the research (Jadhav, 2021).

Schoitsch et al. (2015) claim that the standard does not adequately address security risks. All
systems are assumed to be working correctly, and this is explicitly stated. As a result, the authors
propose a standard that takes risk and hazard into account throughout the analytical phase. In
addition, they suggest that related safeguards be put in place across the whole automobile
engineering process, especially in the area of security. It's no secret that the automobile industry
has been working hard to develop safe and secure autonomous cars in order to reduce the threat
of cybercrime. To combat automotive cybersecurity problems and protect its assets, the sector
has made significant investments in developing standards, according to the study. The
management of automotive cybersecurity has already been the subject of a number of initiatives
and recommendations.

According to Schmittner et al. (2018), among the first initiatives created were EVITA and
EMC2. According to the research, the release of the SAE J3061 Cybersecurity Handbook was
the next step after these initiatives were completed. During this time, engineers studied
automobile cybersecurity from an engineering perspective. Thus, the handbook failed to fulfil the
same function as ISO 26262:2011. Because of this, existing cybersecurity standards like IEC
62443 have to be modified to accommodate automotive engineering processes.

They also say in their study that the automobile sector agreed to create an ISO/SAE 21434
standard for "road vehicles – cybersecurity engineering," according to Schmittner and colleagues
(2018). According to the findings of the research, ISO-SAE 21434 and SAE collaborated to
create a cybersecurity standard for the manufacturing of road vehicles. To guarantee
cybersecurity in the manufacturing of internal vehicle systems, ISO-SAE 21434 outlined a
systematic process. As Schmittner et al. (2018) point out in their paper, this lessens the
likelihood of a possible assault. In addition, the ISO-SAE 21434 provides a way to reliably
respond to cybersecurity threats throughout the globe. A standard for automotive engineering as
well as cybersecurity may be found in the examined literature, and it can be shown to be
necessary. These changes are required since previous initiatives and guidelines neglected to
consider the problem of security. Automotive cybersecurity may be managed using the planned
ISO/SAE 21434 standard. Automotive SPICE may be used in security audits that pay careful
attention to details like password complexity. This provides significant support for the new
ISO/SAE CD 21434. It will have an enhanced and updated methodology shown in this document
(Macher et al., 2020).

The connected Car

The Connected Car is a concept for next-generation cars that has connecting points to external
networks like the Internet. By 2022, there would be 380 million linked vehicles on the road,
according to a study from Business Insider. In order to improve the driving experience for
drivers, these cars feature numerous Internet and external network connections. New technology
and functionalities are available in today's cars that were not previously feasible without even an
Internet connection. Through the vehicle's dedicated cellular connection, passengers may now
get service information including traffic updates. Some new elements of the vehicle's
entertainment centre may be accessed through an Internet connection via Bluetooth or Wi-Fi
when a smart phone is connected to it. With this centre, you can do things like browse the web or
access social networks or stream online material, depending on your car and the manufacturer.
Vehicles will soon be able to interact to each other and to the infrastructure surrounding them
through vehicle-to-vehicle communication (V2V) (V2I). The goal is to make driving safer and
use less gas by giving cars greater control. To name a few key characteristics of this new
technology, they will be able to calculate the speed limit as well as link to the Internet or other
networks to gather road and traffic condition data (Jadhav, 2021).

Vehicle-to-Vehicle (V2V) technology allows wireless communication between vehicles and

even the maintenance of temporary networks between them in order to prevent accidents and
traffic hazards in advance. According to Figure, automobiles will have a network field
surrounding them, and any automobile passing through that field will be able to connect to that
automobile and exchange data. Keeping a suitable distance between two automobiles to avoid
accidents is one use for this data, but it may also be linked to traffic security and safety. The
dispersed nature of this technology makes it difficult to deploy, and in order to do so, automakers
must come to an agreement on the communication technologies and methods to be utilised. As a
result, only automobiles from the same manufacturer will be able to interact with one another.

It is illustrated in Figures 1 that vehicles can interact with their surroundings, including street
signs, motorways, traffic signals, and other traffic-related infrastructure. They can also get up-to-
date data from Traffic Information Centers about potential traffic jams as well as accidents.
Vehicles may cooperate to decrease traffic congestion and prevent accidents by gathering data on
traffic conditions. When all vehicles are linked in the near future, traffic may be managed by a
single authority, providing recommendations on the optimum route and speed to all vehicles,
resulting in an environment free of traffic jams, accidents, and using less fuel. This technology is
already being enabled by a CAR2CAR communication collaboration made up of European
automakers, suppliers, and other research institutes. This consortium includes all of the main
companies in the European automobile sector (Ring et al, 2018).

Automotive networks

Growing electrical components in cars means more demand for a reliable network to link them.
As a result, different electrical components need different kinds of connection to carry out their
various tasks. Automobile networking has evolved into four distinct kinds in the last several
years:

LIN (Local Interconnect Network)

This kind of network is a low-cost way to link switches, intelligent actuators, weather sensors,
tiny motors, lights, and solar roof or heating controls. As the network with the lowest bandwidth,
it is utilised for non-critical tasks of the vehicle's electronic system.

CAN (Controller Area Network)

It is the most commonly used network protocol in the automobile industry. All the data in the car
is transmitted over a single centralised network bus. Depending on the kind of traffic, this
network type may be split into two groups: real-time control in the engine. Electronic throttle
control, anti-lock brakes, and engine timing controls all make use of it (Jadhav, 2021).

Flex-Ray
In order to handle modern drive-by-wire systems, like steer-by-wire and brake-by-wire, this
network type has a high transmission rate need and excellent error control.

MOST (Media Oriented Systems Transport)

Audio, video, navigation, and telecommunications systems rely on MOST, which has the highest
bandwidth of any network. Applications requiring real-time video and audio transmission are
best suited to this technology.

Cyber-attacks on automobile

In 2003, the Embedded Security in Cars (ESCAR) Workshop was established, and research on
automobile cyber-security became more concentrated and open as a result.. In the beginning,
there was little media attention given to this study and consumers did not think it significant.
However, in recent years, cyber-security for cars has received a great deal of attention, mostly as
a result of several very major attacks that have been reported and identified by researchers (Ring
et al, 2018).

BMW

Since the assault on the Jeep Cherokee, and even today, a number of security flaws have been
found in the vehicles of several major OEMs. A security vulnerability in BMW's Connected
Drive service, which allows malevolent users to unlock the vehicle from a distance, was
identified. This flaw affected about 2.2 million vehicles. For the experiment, researchers used a
mobile cellular base station to generate an untrue cell network near the vehicle. A remote unlock
attack was made possible via the car's internal SIM-based phone network rather than the actual
network. As opposed to the genuine BMW server, the vehicle communicated with a spoof of the
latter. Because the communication was not encrypted, the attack was feasible. Eventually, a fix
was released, encrypting the communication between the vehicle and BMW's servers on the
outside using HTTPS.

Tesla S model

Lookout's Kevin Mahaffey, and CloudFlare's Marc Rogers, both mobile security experts,
revealed in 2015 a method to remotely hack the Tesla S and get remote connection to its features.
In this instance, the researchers only had accessibility to the vehicle at initially, but they were
able to get remote access after putting the Trojan into system. Tesla immediately rectified the
problem.

Nissan's Electric Vehicle

Recently, there has been a security problem with a connected automobile using Nissan's LEAF,
one of the world's most popular electric vehicles. Troy Hunt, a security researcher, found a flaw
in the Nissan App, which is installed on smartphones and gives users remote access to many
vehicle functions. The researcher conducted an unusual experiment in which he remotely
attacked a vehicle in the UK from Australia. Air conditioning, seat heating, and the car's GPS
history were all controllable via his smartphone. Despite the fact that the assault was not life-
threatening, it nevertheless serves as a stark illustration of how poorly connected vehicle security
is being handled (Scalas et al, 2019).

Tesla all models

Until now, the only known linked car security problem has been with Tesla. Tencent researchers
found a vulnerability in the Tesla S's Wi-Fi connection, allowing them to remotely control the
vehicle's control system and even apply the brakes. Tesla's web browser, which uses the open
source WebKit foundation, had a vulnerability that gave them access to the vehicle's internal
CAN network. In order to exploit this risk, they created a specially built web site that allowed
them to execute malicious software on the browser when it was visited. Their hot-spot was called
Tesla Guest, which would be a commonly used hot-spot at Tesla dealership shops, and the
password was the same as it was at that Tesla shop in order to present this assault. Tesla vehicles
were immediately connected to their hotspot, which loaded the malware website. A vulnerability
in the Linux system and a flaw in the gateway system, which separates the head unit from
internal CAN network, also enabled the attackers to take control. Tesla issued a security fix ten
days after the breach was made public. In addition, a new security mechanism known as code
signing was implemented, which prohibited firmware upgrades if they were digitally signed by
an encryption module that only Tesla had (Sim et al, 2018).

AUTOSAR

AUTOSAR was established in 2003 and is now the de facto standard among automotive OEMs
as well as their suppliers. The aim was to create an architecture that could be used by the
automobile industry to minimise the growing complexity of software in contemporary cars,
regardless of actual ECU hardware. Because AUTOSAR covers the underlying hardware, the
applications built on top of it are not reliant on the ECU hardware's actual manufacturer. Using
AUTOSAR has been implemented by Volvo Group as the first commercial vehicle
manufacturer. Using this design, the automobile industry will have a consistent standard while
yet having freedom to innovate in terms of new applications and functionality. By reusing
application interfaces and AUTOSAR's fundamental functionalities, adopting the AUTOSAR
standard speeds up the development of new applications. When creating automotive applications,
AUTOSAR creates a standardised template and workflow to guarantee reusability and
dependability.

Using AUTOSAR's standard documentation, automakers may create and deploy software in their
cars with more ease. It's possible for businesses to create software solutions that operate on any
ECU in a car by adhering to the AUTOSAR standard, regardless of the ECU's hardware.
AutoSAR is also known as hardware-independent architecture because to this fact. Other than
the architecture and interfaces, this standard provides a design flow for software mapping to the
ECUs throughout the development cycle (Scalas, 2019). This procedure makes certain that all
businesses utilise the same method to putting this standard within their automobiles. With the
great degree of customisability of the AUTOSAR implementation, any company's application
software may be guided to meet their particular requirements. By using the expertise and
experience from all member companies, this standard is able to better serve its users while also
allowing companies to create their own software applications as well as customise their
AUTOSAR implementations according to their own requirements. This enables businesses to
"cooperate on standards while competing on execution."
 Chapter 3: Research Methodology

When it comes to research methodology, various methods are used to get relevant answers to the
desired question. In research methodology, philosophical and theoretical assumptions behind
research are discussed, as well as the consequences for the technique or procedures that are used.
Methodology is the study of how to do research (Saunders. M, et al, 2007).

"The overall procedure that guides the whole research project" is what researchers refer to as
research methodology. The "main evidence generating mechanism" is another name for
methodology (Prashant et al, 2003). The primary goal in doing this study is to examine the
cybersecurity trends in modern automobile industry. 

As a result, the purpose of this chapter is to outline a methodological framework that will be used
to operationalize the theoretical foundation and address the research goals in order to provide a
more comprehensive understanding. Methodological decisions will be outlined and justified in
detail in this chapter. 

This methodological decision has consequences for both theory and philosophy, which will be
explored in more detail below. It's critical to acquire a knowledge of the research process before
making this methodological decision. There are five levels in Saunders et al, (2016)  "research
onion," which depicts the organisation of the research process. These layers are: "research
philosophy," "research methodologies," "research strategies," and "time horizons."

The research process:

The'research onion' will be used in this study to guarantee that all of the required data is collected
in order to meet the goals. This is due to the fact that doing research is similar to peeling back the
layers of an onion in order to get to the core problem of how to gather the required data to
address the research questions as well as goals, key layers must first be taken away. Using this
method, I can develop a plan for the study's measurements. Below is a schematic of the research
process onion, adapted from Saunders et al (2007), which explains the ideas of each stage of
methodology thoroughly The'research onion' is divided into five stages: theories, methods,
tactics, options, and a time horizon. And my investigation will proceed in accordance with these
phases. Choosing a research philosophy is the first step; deciding on a research methodology is
the second; deciding on the best research strategy is the third; determining the time horizon a
researcher will use for his study is the fourth; and deciding on data collecting techniques is the
fifth.

Research Philosophy

It is essential to emphasise the research's philosophical direction in order to give justification for
the conceptual approach that will guide the investigation. According to Creswell (2013), while
deciding on a study design, the research paradigm is the most critical factor to consider. It's
important since it reveals the researcher's opinion on knowledge creation, which has an impact
on how one does study (Saunders et al., 2012). A researcher's research paradigm determines the
technique of study they will use to collect data.

Positivism

As a philosophy, positivism uses scientific reasoning to build new knowledge (Remenyi et al.,
1998). In other terms, the researcher collects and analyses data. As a result, positivists believe
that reality is a single entity that exists independently of the researcher. This reality may be
objectively assessed, for example by utilising a quantitative survey instrument. Using a scientific
method, positivism aims to enhance knowledge by developing hypotheses and testing them. It
focuses on evidence gained through reducing facts (Gray, 2013).

Interpretivism

This school of thought views the world as a social construction that is also subjective (Bryman
and Bell, 2015). In the philosophy, the significance of phenomena is emphasised, and an attempt
is made to comprehend the situation by being involved in it (Saunders et al., 2009). This means
the researcher is an integrated component of the study process, rather than a separate entity.
Philosophically, science is motivated by human curiosity, and interpretivism's goal is to gain in-
depth knowledge by putting theories into practise (Maylor and Blackmore 2005). As a result, the
philosophy implies subjectivity because the scholar cannot be completely disconnected from
what is being studied and seen.

The research is qualitative in nature therefore, it has focused towards interpretivism philosophy.

Research approach
This decision must be made after considering whether the study should be qualitative,
quantitative, or a combination of the two. In general, researchers must determine what kind of
study is appropriate for a given issue. According to Saunders et al. (2012), the difference
between quantitative and qualitative research is based not just on the issue of quantification, but
also on the context of knowledge as well as objectivity.

This respective study has focused towards using qualitative approach. Qualitative research
collects and analyses textual information. When we talk about textual data, we are referring to
information collected via methods like as surveying, interviews, focus group discussions,
observation, conversation analysis, and exploratory research. The researcher creates a study topic
and specifies that it will be addressed in a qualitative manner. To address the research question
presented in the qualitative study, evidence should be collected (Greene, 2007). To properly
answer the questions, the researcher must provide sufficient information, including specifics,
context, and examples. Qualitative research seems to be the most straightforward of all research
methods. A common mistake when deciding on a research technique is to assume that it will
provide positive results.

It's interesting to do qualitative research since it demands researchers to use dependable data
collecting and analysis techniques to ensure the validity of their results. A researcher's capacity
to come up with a successful research study necessitates meticulous preparation of the research
design, which includes all research elements. It is expected that planning would include
consideration of research variables such as questions and methods for gathering data. Qualitative
research techniques, in general, place a premium on their natural context, which means that they
put a priority on the real-world situations in which phenomena develop. In a natural
environment, a researcher is not allowed to conduct any experiments that may disrupt the flow of
events. In a natural environment, the phenomena may be seen exactly as it is. 

As previously stated, a researcher conducting a qualitative study has a wide range of options for
study design. Qualitative research methods such as observation enable the researcher to see
events in their natural setting. To obtain pertinent papers and records, researchers must use
observation. This gives them access to material that may be crucial to their results and analysis.
In contrast to other research techniques, observation may be more time consuming since the
researcher may get overwhelmed by the sheer volume of information and data available. It is also
possible to conduct interviews in order to gather data for qualitative research. Interviews may be
either official or informal. An unstructured interview allows the researcher and respondents to
have a more in-depth discussion about the subject matter at hand without any restrictions
(Creswell, & Plano 2011). It is possible to obtain further explanation during an unstructured
interview, as long as the researcher asks the right questions. Respondents are also given the
option to provide more information than just a yes or no answer. Also, textual materials are good
data sources for a qualitative researcher. Personal papers, questionnaires, or governmental
records may all be used by a researcher. Meeting minutes, books, peer-reviewed journals, and
secret papers are examples of official documents.

Data collection

In order to collect the data, the research has focused towards secondary data collection.
Secondary research involves interpreting and analysing data that has already been gathered by
another party. Secondary data sources included both raw and summarised reports. Secondary
data may be qualitative or quantitative; descriptive or explanatory; secondary data can be utilised
equally. It's simple to create a research project using secondary data that's based on other
people's business expertise and experience. In contrast to collecting primary data, secondary data
may be acquired quickly and at a lower cost (Zikmund, 2003). Secondary data, on the other
hand, has the drawback of not being tailored to the researcher's requirements and the user having
no control over the correctness of the data.

Secondary data has been classified in various ways by various scholars. These categorizations,
on the other hand, do not cover the whole spectrum of information. Secondary data is divided
into three major categories: documentary data, survey-based data, and data collected by other
researchers (Saunders, 2007). It is from a variety of secondary sources that I am compiling my
study material. Secondary sources have the advantage of providing an interpretation of data
gleaned from main sources. Think about things like dissertations, databases, summaries,
references and journal articles as well as books, newspapers and the internet.

Research strategy

Methodology refers to the approach used to identify, analyse, and answer research questions in
order to meet the study's goals (Remenyi et al., 2005). It is possible to do research more
methodically when using a strategy, which is an action plan that addresses research issues and
problems. Surveys, experimentation, action research, deductive research approach, case study,
and exploratory research are some of the frequently utilised techniques (Saunders et al., 2016).

Data analysis

In order to conduct this research, the researcher has focused towards creating and develop
themes with the help of content and statements of previous researchers. The study will link the
statements with each other to generate a theme. Therefore, the respective research has focused
towards thematic analysis that how trends in cyber security impacting modern automobile
industry.
 Chapter 4: Analysis

Because of advancements in the fields of artificial intelligence (AI) as well as machine learning
(ML), the use of autonomous systems has grown in recent years. The integration of autonomous
systems into our life will occur in practically every sector in the near future, including smart
home devices, delivery drones, as well as driverless autos, among others. These technical
improvements in the car sector, on the other hand, have resulted in significant cybersecurity
concerns. If thieves are successful in hacking or controlling such devices by exploiting their
weaknesses, computers betraying humans will no longer be the stuff of Hollywood movies
(Amin, and Tariq, 2015). Driverless automobiles, unmanned aerial vehicles (UAVs), and the
Internet of Autonomous Things (IoAT)/Autonomous Internet of Things (IoAT/AIoT) are three of
the most significant advancements in the automotive sector as they relate to cybersecurity
(AIoT). It will be the emphasis of this study to examine the technical growth of autonomous
systems as well as their vulnerabilities to assaults

Trends in automotive industry

The era of autonomous cars has already started, and it will continue. The self-driving technology
has been accepted by major automobile firms like as Tesla, Uber, and Waymo (McBride, 2018).
Other automobile manufacturers, including as Google, Volkswagen, BMW, Ford, as well as
Nissan, have all envisioned a future in which autonomous cars are commonplace on the road. It
is projected that driverless autos would reduce traffic accidents, road congestion, fuel usage, and
air pollution in the future. Modern automobiles are equipped with advanced driver assistance
functions like as automation, self-parking, as well as summoning, as well as blind spot and lane-
monitoring systems, among others (Checkoway et al., 2011). Despite this, there are still
significant difficulties to overcome in order to make cars with these sophisticated autonomous
capabilities totally driverless and safe on the road. In Arizona, for example, a self-driving Uber
car was involved in a deadly accident that killed the driver (Bergin, 2015).

Vehicle Ad hoc Network (VANET) is a system of automobiles that are linked to wireless
networks via the use of cellular technology. Future autonomous cars will need to be linked to one
another via highly developed vehicular communications such as VANET in order to
communicate traffic as well as routine info such as speed, position, and notice of any traffic
collisions with other vehicles (Clark et al., 2017). This, on the other hand, raises the likelihood of
cyberattacks. Earlier this year, Wired published the results of a hacking operation on a Jeep
Cherokee in which an attacker hacked the vehicle and took control away from the driver while
driving on the freeway. When the accelerator quit functioning, the driver's experience got tense
since there was nothing to lean against for safety. Taking control of brakes and disabling them
completely would have made things far worse. Vehicles' driving judgments and performance
would be improved if they could communicate more effectively via VANET.

However, a considerable quantity of private information would be circulated over the network,
increasing the likelihood of cyberattacks. Cyber attackers would have the opportunity to listen in
on a user's commute habits, which may disclose a great deal of important information about the
individual. As a consequence, cyberattacks might be carried out while the victims are on their
way to work. As an added precaution, cyber criminals might deliver erroneous congestion alerts
to automobiles by taking advantage of the VANET's congestion avoidance algorithms. This
would provide attackers the ability to reroute and position themselves in topics of focus (Cui et
al., 2018). It is essential to investigate the infrastructure and primary features of the area, as well
as to identify and assess the appropriate defences against potential vehicle assaults. VANETs are
vulnerable to cyberattacks, according to researchers, and identifying and categorising these
vulnerabilities and security problems might help limit the amount of intrusions. Unmanned
Aerial Vehicles (UAVs) are aircraft that operate without the presence of a human pilot.

Commercial unmanned aerial vehicles (UAVs), such as drones, are becoming more popular as
more and more companies seek to employ them for commercial reasons. UAVs may be
employed for a variety of purposes, including deliveries, ground research, and military tactics.
For instance, Bell Labs is working on prototypes that can transform into an aircraft in the middle
of a flight while utilising either gas or electricity (Pappalardo, 2018). The prototype is utilised for
delivery, which reduces the number of wires, costs, and time required during the deployment
procedure. These systems, on the other hand, are very vulnerable to cyber attacks. The two most
popular types of assaults against UAV navigation are jamming the Global Positioning System
(GPS) and spoofing the GPS signal (Macher et al, 2020). GPS jammer is the technique of
interfering with GPS signals by employing a device that transmits at a certain frequency. The
goal of a GPS spoofing attack is to broadcast a phoney GPS signal that is received by the GPS
receiver of autonomous system, which then calculates a fictitious position, navigational, and time
(PNT). Consequently, the contradictory data leads to the system deviating from its original
course, maybe toward a malicious destination, or to performing other wrong system activities.
When the US Maritime Administration submitted a report on GPS spoofing, it revealed that
roughly twenty vessels sailing off the Russian port of Novorossiysk ended up in the incorrect
location more than 32 kilometres inland, near Gelendzhik Airport. Other incidents have been
reported (Hambling, 2017).

While military GPS transmissions are encrypted, normal citizen GPS signals are not. Regular
citizen GPS signals is publicly available. When GPS jamming & spoofing assaults are carried
out maliciously, they pose a substantial danger to key infrastructure and the general public's
welfare. From the researchers' point of view, there are 2 different notions that are connected. One
example is the Internet of Autonomous Things (IoAT), which is a newly coined word for
technology that relies on autonomous devices to function without the intervention of humans.
These gadgets would be linked together via a network and will be able to solve issues or adapt to
their surroundings by exchanging information with people and other machines in their
environment. In addition, there is the Autonomous Internet of Things (AIoT), which is a
collection of embedded devices that actively manage data and make choices on their customers'
behalf ). These two ideas are related in that they both include intelligent gadgets that make
independent choices while remaining linked to one another over a network (Obaid et al, 2020).

Future developments will include autonomous gadgets that will communicate with cloud-based
control systems. Cloud computing, machine learning, as well as blockchain technologies are
expected to advance in tandem with the growth of Internet of Things devices, and it is projected
that these machines will achieve autonomy on par with human actors in the near future, if not
sooner. Despite the progress of IoAT & AIoT, there remains a paucity of study in this field.
When it comes to exploring these issues, one of the most important things to consider is
cybersecurity and data privacy. Similarly, a list of cyberattacks that may be carried out on IoT
devices might be useful for AIoT devices as well.

Embedded Devices

Devices embedded in our increasingly networked and data-centric society play a crucial role,
particularly in critical infrastructure (e.g. intelligent grids), smart cities, as well as IoT
architecture (Bertino et al., 2016). Embedded systems in automobiles have become more
complicated. These gadgets, on the other hand, often operate on a limited electrical budget and in
an unpredictable (and sometimes hostile) environment with limited connection. Testing and
debugging industrial software systems as well as cyber-physical systems, such as those, may thus
be a tough procedure. It's difficult to deal with recurring defects, for example, when bug reports
are regularly insufficient (Hu et al., 2016). Systems that are resource limited and tightly
integrated with their operating conditions provide limited support for sophisticated debugging
tools. Smart automobiles (such as Tesla smart cars, for example) are gaining popularity as time
passes. A variety of on-board and embedded electronics are used in these vehicles. Examples
include on-board processors, sensor, transceivers, and storage. Embedded device innovation and
cyber manipulation of these devices as attack vectors will continue (e.g., vulnerabilities). Secure
data exchanges between vehicles and underpinning infrastructure are critical to avoiding
automotive failures that might result in fatalities (Choo et al., 2016).

Prevention of cyber attacks on Autonomous Systems

When it comes to cyber-physical systems with embedded devices linked to IP and managed by
one or even more network operators, autonomous systems fall within the category of cyber-
physical systems. The protection of these systems is essential since they are susceptible to
cyberattacks. When these systems go down as a result of cyberattacks, the user may be left
perplexed or uninformed (Sabatier et al., 2014). IT workers and researchers can discover and
analyse cyber threats and vulnerabilities in a system, then develop remedies to minimise the
hazards. IT personnel and researchers can help avoid cyber attacks. In addition, businesses
should undertake frequent security upgrades to guarantee that the network and data of their users
are safe. Malware activity, network assaults, and strange instructions may all be detected using
machine learning techniques, which can be used to create behavioural profiles of each possible
attacker (Critchley, 2018). By constantly updating and being free of any human or technological
objectives, this would also increase the efficiency of the cryptographic algorithms (Sabatier et
al., 2014)

Wireless connection technologies are being used in an increasing number of cars, and completely
autonomous vehicles will eventually hit the market, which is both worrying and exciting. A high-
speed wireless network with roughly two million PCs is ripe for hackers. By providing
misleading location and speed data, a hacked autonomous vehicle risks confounding other
vehicles on the road as to where the infiltrated vehicle is. As a result, they may be forced into the
median or another car. Assaults like as jamming and denial-of-service (DoS) attacks might be
used to prevent an automobile from receiving the data from other cars or to disrupt regular
functioning while the vehicle's computer tries to handle the influx of communications (Linkov et
al., 2019). This has the potential to be very harmful to both individuals and the environment,
which is why it is imperative that the possible dangers be assessed and the automotive industry's
cybersecurity be strengthened.

Threats to cybersecurity of an automobile

Attacker model

One may construct the attacker model by using the numerous examples discussed in the
preceding sections. The attacker in the attack scenario has both physical as well as wireless
connection to the vehicle, as seen in the image below. To attempt to retrieve the communication,
the attacker may listen in on internal communication, acquire it, replay it, alter it, or inject it into
the untrusted channel. This model is based on the Dolev-Yao assailant model. By exploiting
these vulnerabilities, a car's security might be jeopardised. It's critical to distinguish between
automobile security problems and general safety concerns. Computer security events occur as a
result of malicious action, as opposed to safety concerns, which occur at random (Shahabadkar et
al, 2012). 

Car Owners

The main group of people interested in changing with the firmware is car owners. Despite their
size, they have a limited amount of information and resources at their disposal. The majority of
automobile owners aren't knowledgeable enough to launch a preemptive assault on their vehicle's
computer system. On the other hand, most individuals are solely concerned with improving their
engine's output, thus they seek for ways to acquire more features or performance of their car
without investing a lot of time or money. It's not implausible to think that paying a premium in
the future may get access to more automotive functions. Purchasing aftermarket auto parts and
accessories has never been easier thanks to current (mobile) payment methods. Modern
automobiles are equipped with subscription-based amenities such as voice calling, GPS and
remote diagnostics.

General Motors' OnStar, Chevrolet's MyLink, and BMW Assist are just a few examples of
subscription-based services. In the future, features such as better steering, greater engine
performance, and more intelligent brakes may be available for a fee or as part of a subscription.
For example, a firmware upgrade on numerous ECUs may make these functions available. The
techniques used to identify a user on a backend portal have varied widely. The VIN, a user
account, a car's mobile phone number, a public/private keypair, or a combination of these
techniques may all be used for identification and authentication. Authentication and
identification methods may be circumvented when an owner of the car wants more material or
features of his vehicle but does not want to pay for them. Take, for example, the practise of using
the same user profile in different vehicles, each with increased rights (Straub, 2017).

Engine Tuners

Car tuning entails making changes to the vehicle's look or performance. Most cars arrive from
the factory pre-configured for the needs of the typical driver, however automobile tuning allows
owners to change the attributes of their vehicle to suit their tastes. It is possible to improve the
fuel efficiency, power, or handling of a car by modifications. Engine tuning is possible because
to the inclusion of ECUs in automobiles. ECUs may be used to improve vehicle performance in
two ways. One method is chip tuning, in which performance chips are installed between engines
as well as actuators, modifying the data received by ECUs and changing the engine's
performance as a result. ECU remapping is the process of altering the engine's software, either
with custom-written firmware or with firmware provided by the manufacturer. Both methods are
used to get the best results, to boost the engine's production, efficiency, or longevity. Increased
performance may reduce engine life because of the additional stress it puts on the engine's
components. ECU remapping has grown in popularity over chip tuning since the mandatory
OBD-II port was mandated in 1986 (Straub, 2017).

Engine tuning has been a profitable business since electronic control units (ECUs) were installed
in automobiles. Almost all current automobiles can have their ECUs remapped. It's common
practise to desolder a chip before removing the firmware and analysing it. A car's performance is
then fine-tuned using firmware settings that have been evaluated. Once the new firmware has
been flashed to the ECU, the vehicle is turned on and the OBD-II port is plugged in. As soon as a
performance or modification pack is released for a certain automobile model, it may be applied
to all of that models (Villasenor, 2011). The prestige that comes with being the first one to
design a flash pack for the a new model of automobile may be worth as much as the money, thus
car tuners are motivated by more than simply profit. Modern ECU tuning is getting more
difficult as automakers and engine tuners engage in an arms race. Modern ECUs have so-called
anti-tuning protection, which means attacks are getting more sophisticated. To properly remove,
change, and re-insert firmware, engine tuners must have in-depth understanding of automobiles,
their communication buses, and ECUs. For this reason, groups of motor tuners will have to
devote a significant amount of time and money to developing a tuning pack.

Security Researchers

When security researchers attempt to access, change, or disrupt firmware, they have ulterior
motives. There are normally well-funded, competent and independent security researchers or
academic researchers; in other words, there is no bias toward any entity. Businesses frequently
get helpful information through security research regarding their goods as a result of their
researchers' work. For example, General Motors was warned by the study team evaluating the
anonymous sedan before their findings were published, which revealed several weaknesses in the
software used in GM's vehicles. The desire to learn newer things and put hypotheses to the test
motivates security researchers. Finally, from the manufacturer's perspective, they provide an
excellent service by identifying possible vulnerabilities and exploits, generally free of charge.
However, security researchers sometimes sell their discoveries on the zero-day market as a result
of various conflicts with the revelation of vulnerabilities (Scalas, and Giacinto, 2019).

Black Hat Hackers

Cybercriminals are classified as either white hat or black hat hackers depending on their
motivation. Black hat hackers do not inform the general audience or the manufacturer about
vulnerabilities they have discovered. For personal benefit, black hat hackers could want to get
their hands on automobile firmware so they can market it to federal agencies, terrorists, or rival
automakers, or so they can use it to create exploits or do other bad things. It's possible to make a
lot of money by selling these weaknesses as "zero-days" Patches or backdoors may be applied to
vehicle systems by black hat hackers after they have taken control.
Ethical Hackers

It is important to remember that ethical hackers follow the same guidelines as security
researchers but have access to additional tools like social engineering that may be used to
compromise a system. For example, ethical hackers may test safety unannounced and
responsibly reveal the flaws they find. When compared to cyber attackers, ethical hackers aim to
make the world a better place instead than making money. A hacker operating on behalf of a
company might have access to a large amount of resources and time. Their curiosity and desire
to expand their knowledge pushes them to look for weak points. Hackers that practise ethical
hacking could seek for automotive firmware to recreate and test for security flaws, or see
whether the firmware can even be obtained (Sabatier et al, 2014).

Theft of Automobiles

Car thieves are motivated only by self-interest and lack the skills or resources necessary to
successfully steal a vehicle, for example. Professional criminal syndicates, on the other hand,
often have an abundance of financial and human resources at their disposal. In general, the
exploits used by these groups will be developed by black hat hackers and marketed to the public.
By connecting a device to an open OBD-II port, vehicle thieves might equip themselves with
tools to remotely unlock automobiles or disable the alarm or GPS. Car thieves who work at valet
parking facilities could target costly cars, park them, and then modify the firmware by putting a
device into the OBD-II connection, making it possible to follow the vehicle and remotely unlock
it later.

Competing Automobile Manufacturers

Automobile manufacturers have whole research departments devoted to fine-tuning their

vehicles' handling characteristics. Resources and expertise are readily available for assaults like
desoldering circuits or man-in-the middle attacks to get access to a competitor's car's firmware.
Firmware like this may provide the business insight into how the firmware handles things like
fuel management, engine optimization, and brake pedal sensing (Linkov et al, 2019). Automobile
manufacturers can't use such tactics because of legal difficulties. Unofficially, though, given
recent problems in the car business, this wouldn't be shocking.
 Chapter 5: Case Base Analysis

Case 1: Ford Escape and Toyota Prius

The cybersecurity of two popular contemporary automobiles, the Ford Escape as well as Toyota
Prius, was extensively analysed. Based on earlier studies, the researchers developed new theories
on how an attacker would try to alter a vehicle's behaviour following a successful assault. These
assaults would allow for the execution of arbitrary remote code on the ECU via one of the
vulnerable channels listed above. Research shows researchers can manage steering, brakes,
accelerating and displaying features on two test cars in specific. Researchers. For the most part,
they went into considerable length in their explanations, providing the source code as well as a
description of the gear required to carry out the assaults. It was not simple to interpret, observe,
and reverse engineer the code and data delivered over the CAN buses even though the
researchers had complete access to all ECUs (Miesel, 2006).

According to the Ford Escape's CAN packet, there was an indicator for how hard the accelerators
was depressed in the bytes of the packet. To aid the ABS figure out whether a traction control
event was taking place, this data packet was transmitted from the PCM. It has nothing to do with
how fast the automobile should go. For instance, packets that show how hard the brake is pushed
but do not apply the brake when replayed were several instances of this. It would need additional
investigation to discover the true meaning of all texts. This was anticipated, given that a
contemporary automobile has millions of lines running it. Another issue is that the recipient
ECUs may be equipped with safety mechanisms that prevent them from seeing the packages the
researchers were trying to send them to begin with. On the Toyota Prius, for example, the
packets used to turn the wheel in Smart Park Assist only operate in reverse. In the same way, if
the steering wheel is turned more than 5%, Lane Keep Assist packets are ignored. The ECU can
be manipulated to get beyond these limitations, but this would involve more effort, making these
assaults more difficult to carry out (Miesel, 2006).

Case 2: Chrysler and Uconnect

1.4 million Chrysler cars were recently recalled after a software issue was discovered by two
security experts. As previously reported, researchers that looked into the Ford Escape as well as
Toyota Prius found various assaults that compromised the security of a Jeep Cherokee and, thus,
a driver's safety, employing remote vulnerabilities in their work published by Miller et al. The
researchers chose the 2014 Jeep Cherokee as their target. Interestingly, both CAN buses are
integrated in this car's radio system as well, including phone, Bluetooth, and WiFi channels. If
the radio is compromised, all of the vehicle's electronic control units (ECUs) may be accessed.
The 2014 Jeep Cherokee's infotainment system, Wi-Fi, GPS, applications, and cellular
connectivity are all provided by Harman Kardon's Uconnect radio (Gottehrer, 2015).

Various vehicle types may use this technology, which contains a system on a chip (SoC) to
supply all of its functions. One of the Uconnect chips communicates with two CAN buses, which
are known as the CAN-IHS and the CAN-C respectively, which stand for controller area
networks. According to the experts, the Uconnect system has a number of problems. It was
possible to reverse engineer WiFi by examining the WiFi interface, that allows users to connect
to a vehicle WiFi hotspot. The E:generateRandomAsciiKey() technique produces a WPA2 WiFi
communication channel password using a random ASCII string. It was possible to reduce the
number of possible passwords to roughly 15 million using this approach, which takes time into
account while generating them. For security reasons, the WPA2 password creation method
utilised the date and time of 1 January, 2013, 00:00 as an input if no valid time could be acquired
before commencing. From there, researchers might easily locate the right WPA2 password by
narrowing down the password selections.

Due to the fact that this hack only functioned while in close proximity to an automobile,
researchers investigated the possibility of hacking into the mobile communication network
instead. It was feasible to connect a computer linked to Sprint 3's cellular network to an open
port on the Jeep thanks to the network's interconnectability. The researchers also discovered the
IP address range used by the Jeep on the Sprint network and discovered 2700 devices with a
simple scan. They were able to transmit instructions to the CAN bus after opening a shell in the
TCU's operating system. The researcher was able to remotely disable the brakes as well as the
engine by sending a few signals. Sprint network and Chrysler addressed the problem within days
of the paper's publication. The vulnerable port had been closed by Sprint and Chrysler. This was
followed by an independent recall by Chrysler for 1.4 million cars (Gottehrer, 2015).

Case 3: General Motors’ Onstar

A researcher found a flaw in General Motors' (GM) OnStar telematics system shortly after the
Chrysler attack. The researcher constructed a gadget called OnStar after listening in on
communication between a smartphone applications called Remote Link, which allows drivers to
manage functions like locking doors as well as turning on lights in their cars, and the cars. The
bandwidths of the automobile receiving an open door order was blocked, but the information was
saved as well. A jammed signal would prevent the automobile from reading it, so the user would
have to attempt opening it again. As a result of this second signal being saved, the OnStar gadget
would play back the initial signal to allow the user to get inside the vehicle. The researcher then
able to operate as if he possessed the automobile using the second signal's information,
determining its precise position, opening the doors, and even turning the ignition key (Farris et
al, 2003).

Even before the study was released, Chrysler voluntarily pulled 1.4 million cars from service and
patched the problem, while General Motors got a similar research article in 2010 warning that
their OnStar telematics system was susceptible to an attack before the research had been
published. It took General Motors five years to address the problem and provide a fix. The only
difference between the two situations was that the researchers in the second one did not specify
the automobile manufacturer or model they were looking into publicly. The 2009 Chevrolet
Impala was the nameless automobile in the two experimental studies discussed previously in this
section. So, it's no surprise that it took GM almost five years to secure its vehicles against the
breach that was confidentially revealed to GM. In this assault, the OnStar dashboards computer
was exploited remotely and was capable of monitoring automobiles, activating the brakes at high
speeds, and even disabling them completely. It's important to keep in mind that the problem here
wasn't with the vehicle or TCU, but rather with the application. The program used SSL to
interact with the GM servers, however the authenticity of the server's certificate was not
validated. Intercepting the transmission and spoofing a GM server revealed that the data was not
encrypted except for the SSL connection. As a result, the researcher was able to get more access
to the program, which allowed him to breach the vehicle (Fitzgerald, 2016).

Case 4: Tesla’s Model S

Among the many technological components of Tesla's Model S is a 17-inch touch screen display
that allows the driver to operate many operations of the car, such as controlling media or
accessing navigation, as well as enabling autopilot and change the vehicle's height. Users must
create a Tesla profile on the company's website. For example, customers may use this account to
access the Tesla ios app and operate their vehicle's functionalities from their phone. The iPhone
app's capabilities include locking or unlocking the vehicle, finding it, and controlling the top of
the car. Tesla's account, according to a 2014 report by Dhanjani, may be vulnerable to hacking
owing to a few factors. In the first place, a user account password had to be at least six characters
long and include both a number and a letter. A brute-force attack might exploit this need since
authentication is single-factor and there seems to be no password protection policy after a set
number of erroneous login attempts on the Tesla website (Endsley, 2017).

Tesla increased the password length requirement to eight characters as a result of the article's
publication. Second, the Tesla REST API allows third-party apps to utilise the user's credentials
to access the REST API on their behalf. Thus, unscrupulous third-party programs might
potentially obtain and misuse Tesla account details. Tesla users were cautioned not to use third-
party apps until an SDK was released by the company. SSH, HTTP, NFS, and telnet were all
visible through the Model S's WiFi connection, which indicated open ports. Nevertheless, since
the researcher didn't go on to learn about these services, they're now at risk of being attacked.

Case 5: BMW’s ConnectedDrive

BMW's ConnectedDrive technology includes features like as remote services, personal support,
and real-time traffic information. A control unit, sometimes known as a Combox, houses the
technology, which has been found in a number of BMW cars since 2010. Using a USB stick or
Bluetooth device, the Combox can play audio files and connect to a modem via
GSM/GPRS/EDGE. In response to an ADAC inquiry into the privacy implications of this
device, a researcher examined the security flaws detected in the control unit. The researcher's
findings were released. At first, this gadget looked to be safe, but a closer look revealed
vulnerabilities. To begin with, the device's emergency services were encrypted using methods
that are widely recognised to be insecure, such as DES (Rigoll, 2007).

The researcher also discovered that BMW utilised the same basic cryptography keys for all
vehicles to interact with the back-end servers and that the secret keys could be readily recovered
from the firmware after dumping and analysing the modem's software. Additionally, certain
ConnectedDrive services from BMW do not encrypt communications sent between the vehicle
and the server at the back end. As a last point, since the messages delivered aren't salted, they're
vulnerable to replay assaults. The discovered flaws have been validated and corrected, according
to BMW. Although ConnectedDrive technology cannot be turned off by the owner since it is
built into the vehicle, a proper written request and a trip to the mechanic are all that is required to
deactivate it. These flaws impacted almost 2.2 million automobiles throughout the globe.

Recommended models and solution

Additionally, there are a number of organisations that have made a commitment to addressing
global security challenges related to automotive systems and finding ways to make such systems
more secure. Several of these projects are discussed in further detail below.

Automotive Open System Architecture (AUTOSAR)

In the electronics, semiconductor, and software industries, AUTOSAR (AUTomotive Open

System ARchitecture) is a global development cooperation of vehicle makers, suppliers, and
other firms. Using AUTOSAR's unique electronic technologies, automakers have improved
performance, safety, and environmental friendliness since 2003. Aiming to remain ahead of the
curve in terms of new technologies while also reducing costs without sacrificing quality,
AUTOSAR is already working for industry standards. Software and hardware updates are also
made easier because of this feature. More than 180 companies, including the Bmw, Bosch
GmbH, Volkswagen, Mercedes benz, Toyota, as well as the General Motors, are integral to the
alliance. The AUTOSAR standards developed by the AUTOSAR Development Partnership may
only be used for free by firms who have joined the partnership (Fennel et al, 2006).

AUTOSAR has published a number of guidelines and recommendations relating to the

performance, safety, and environmentally friendly nature of automotive systems, but none of
them address the issue of firmware security. Additionally, the AUTOSAR stack design consumes
a significant amount of additional resources, which poses concerns with resource-constrained
ECUs. Due to the automobile industry's focus on cost-cutting, manufacturers often compromise
on quality. Additionally, since the requirements are interpretive, the majority of participants have
their own private implementation of these specifications, which is the opposite of the
partnership's objective.

E-Safety Vehicle Intrusion Protected Applications (EVITA)

The goal of the E-Safety Vehicle Intrusion Protected Applications (EVITA) project was to
design, verify, and prototype a layout for automotive on-board networks that protects security-
relevant components from tampering and sensitive data from compromise while being
transferred inside a vehicle. A consortium including Bmw Group, Porsche Research, and
Infineon managed the project. From July 2008 to December 2011, the project was active. Several
security criteria for car on-board networks were determined when the team began the project by
looking at the stated use scenarios for automotive networks.. The next step was to create, test,
and build an infrastructure for the on-board automobile network. EVITA's security duties are
separated between software and hardware, with hardware encryption modules (HSMs) serving as
the root of trust rather than the ECUs themselves. Cryptographic coprocessors were added to
ordinary ECUs using Field-programmable circuit arrays (FPGAs) during prototyping (Fürst et al,
2009).

The secured on-board architecture and communication protocols have been designed using UML
and evaluated with model-based verification tools to guarantee that the defined criteria are met. 
The architecture and protocols that resulted were released as open standards. Custom AUTOSAR
specs are used in certain of the requirements. EVITA is divided into three categories; full,
medium, and light. In important applications, such as the car's head unit, which handles
communication with the outside world, EVITA Full offers security. AES-128, Whirlpool, as well
as ECC-256 are among the cryptographic primitives supported by EVITA Full. EVITA Medium
may be utilised for less essential automotive applications, such as communication between
ECUs. The ECC-256 and WHIRLPOOL cryptographic primitives are omitted from the Medium
HSM to let it work with less powerful ECUs. Only AES-128 symmetric encryption is supported
by EVITA Light, which is more often found in sensors and actuators. For example, pre-
configuration during manufacture, self-initialization or executing a key establishment procedure
in software at the associated application processor may all be used to create the requisite shared
secret.

Audi, BMW, Daimler AG, Porsche, and Volkswagen are all members of the Herstellerinitiative
Software (HIS) consortium, which stands for "OEM software initiative" in German. The project's
goal is to make software in ECUs as uniform as possible. Software packages, program tests,
software solutions, and bootloaders are only a few of the projects undertaken by the consortium's
several groups. Efforts have been made to speed up flash programming while also improving the
device's overall stability. Firmware flashers are created using protocols described in the standard.
Despite the fact that HIS does not directly address security issues, the concepts and standards
found in HIS may be used to this project.

Selected model and its implementation from threats of cyber security

AUTOSAR

Since its inception in 2003, AUTOSAR (Automotive Open System Architecture) has been the de
facto industry standard for automotive original equipment manufacturers and their suppliers. As
a result, researchers set out to create an architecture that could be used by automakers to decrease
the growing complexity of vehicle software. Because AUTOSAR abstracts the underlying
hardware, the applications built on top of it are not reliant on the ECU hardware's original
manufacturer. Volvo Group introduced AUTOSAR to the commercial vehicle industry. Using
this design, the automobile industry will have a common standard while yet having space to
innovate in terms of new applications and functionality. It saves time to use the AUTOSAR
standard because it allows developers to reuse existing application interfaces and key
AUTOSAR functionality when creating new ones. When designing automotive applications,
AUTOSAR uses a defined template and work flow to assure reusability and dependability
(Heinecke et al, 2004).

Architecture of software

Using AUTOSAR's standard documentation, automakers may create and integrate software in
their cars with more ease. It's possible for firms to create software solutions that operate on any
ECU in a car by adhering to the AUTOSAR standard, regardless of the ECU's hardware. That's
why it's referred to as a hardware-independent architecture on the AUTOSAR platform. This
standard not only describes the architecture and interfaces, but it also provides a design flow for
how software must be mapped to ECUs throughout the development cycle. In this way, the
standard is implemented uniformly across all car manufacturers. With the great degree of
customisability of the AUTOSAR implementation, any company's application software may be
guided to meet their individual demands (Fürst et al, 2009). To achieve this standard, one aim is
to leverage the expertise and experience of each member firm to better and then further develop
the standard while at the same time allowing each company to create its own application
software and customise the AUTOSAR implementations to meet their own requirements.. The
slogan of AUTOSAR is “Cooperate on standards, compete on implementation.”

Security features

It covers crucial security techniques used by SWCs as well as other software modules integrated
into the car's system, which are defined by the AUTOSAR standard On-board communication
security is ensured by interfaces and processes specified in the standard, however OEMs are free
to decide on other security methods and implementations. Custom equipment manufacturers
(OEMs) are in charge of selecting the cryptographic methods and encryption methods that will
be used in their vehicles. Crypto Service Manager, Crypto Abstraction Library, and Secure On-
Board Communication make up the AUTOSAR's three core security mechanisms.

CSM and CAL

AUTOSAR calls for the use of two different cryptographic modules. Firstly, we have CSM,
which is positioned in the BSW's Service layer and serves higher-level applications with
cryptographic services. Using CSM, various applications may access cryptographic primitives by
using the same service (cryptographic algorithms). In certain cases, one program may need an
MD5 authentication while another will require a SHA1 digest to be computed. Because
AUTOSAR does not specify access across ECUs, all access to the CSM module's service must
be described and provided as a separate method (Fennel et al, 2006). A Crypto Abstraction Layer
(CAL) is a static library that serves a similar purpose to the CSM as the second crypto module.
The library bypasses the run-time environment to deliver cryptographic functions directly (RTE).
Other software packages including BSW, Software components (SWCs), or the Complex Drivers
layer may call CAL's C-functions directly. Since the CAL's functions are constantly called, they
are always performed as a function call. There is no connection between CAL, which is a library,
and any of the AUTOSAR architectural levels. No cryptographic methods are defined by CSM
or CAL; instead, implementers choose which algorithms to use depending on their preferences or
client requirements. For this reason, it is critical that OEM security employees have familiarity
with a variety of cryptographic algorithms and know which ones are safe to use. Both modules
offer a cryptographic algorithm selection input parameter for a software component or module.
The CSM or CAL may implement the following cryptographic functions (Fürst et al, 2009).
Secure On-board Communication (SecOC)

The SecOC module protects vital data by authenticating users. It's found in all electronic control
units (ECUs) that need to communicate securely. The AUTOSAR trusted system first mentions
this module in Release 4.2. Using the module's security mechanism, current communication
technology may be easily secured without using a lot of resources. Asymmetric authentication
procedures are supported as well, however the module's design assumes that symmetric
approaches would be used using Message Authentication Code (MAC). Symmetric
authentication is quicker, simpler, and more secure than asymmetric authentication since it uses
smaller authentication keys (Fennel et al, 2006).

The AUTOSAR Security Work Package

All automotive security professionals, including vehicle manufacturers, distributors, software

stack providers and semiconductor vendors, joined the AUTOSAR Cybersecurity Work Package
in November 2014. The goal of this project is to meet the global security issues in the automobile
sector, particularly with regard to the linked vehicle idea and any resulting security risks. Higher
quality security measures may be developed with the help of additional specialists, and these
measures will then be integrated into the AUTOSAR standard. Further benefits include lower
costs and better interoperability among various ECU manufacturers. / Because the firms aren't
working on the solution independently, the expenses are lower. Instead, they collaborate and
pool their resources to discover the best answer more quickly and efficiently (Fürst et al, 2009).

AUTOSAR's Prospects for the Future

The AUTOSAR organization's key future product is the AUTOSAR Adaptive platform, due in
early 2017. The AUTOSAR partners may create the first execution of the AUTOSAR standard.
Till now, the AUTOSAR was built on detailed specification documents, which each partner used
to build their own execution of the software. With the Adaptive platform, however, each partner
will have access to software that has already been built and is licenced to all AUTOSAR partners
as standard. The AUTOSAR group hopes that its new Adaptive platform idea will meet the
needs of future cars, including improved connectivity, self-driving capabilities, and a more
dynamic design that interacts with the other vehicles as well as the surrounding environment.
IP/Ethernet technology will be used for communication on the Adaptive platform. It will have
just the most basic operations, with no security modules, and it will grow in scope with each
subsequent version as new features & modules are added. As far as the programming language
goes, it'll be in C++. It's anticipated that security measures such as encrypting data and crypto-
hardware would be included. By adding extra features and shortening development times, the
Adaptive platform complements the Classic platform rather than substituting for it.

Summary

AUTOSAR has a lot of promise, and it's already being used as the foundation for a lot of car
software. AUTOSAR's security mechanisms and practises will be incorporated into all cars that
utilise them if they adopt strong security mechanisms and practises, and vice versa. There is a
very minimal danger of being abused by bad attackers if we include the individual security
procedures that each business puts to its automobiles. The existing security components in
AUTOSAR must be highlighted, but the standard must also be tested to identify what kinds of
security upgrades may and should be implemented.
 Chapter 5 - Conclusion

Summary of the study

As the vehicle industry grows more aware of the hazards presented by cyber assaults, research
into the security flaws in this standard is becoming more extensive. As a result, any research that
discovers possible vulnerabilities today may not be implemented in vehicles until 2020. The
problem is that new automobiles for the next 2-3 years are basically "completed" but, certain
software may need to be updated via service centres, all programs adhere to the AUTOSAR
specification, and the vehicle's operating system is created in compliance with the AUTOSAR
specification. Although we may be successful in upgrading the software, each application that
was previously running on the old operating system will need to be updated in order to work
correctly on the newer version of the software. There's also the matter of the data kept in the
automobiles, which is necessary in order for them to function effectively. If you want to keep all
of your programmes and data running properly, moving from Windows 7 to Windows 10 may be
a challenging task. However, doing it in a vehicle that wasn't even meant to provide this as an
option is very hard to do (Villasenor, 2011).

Cybersecurity refers to a combination of technologies, processes, and actions that are designed to
prevent hackers from obtaining access to a computer system without the proper authorisation.
The relevance of cybersecurity has increased significantly in recent years. As per Schmittner and
Macher's analysis in 2019, a large number of self-driving automobiles are connected together.
Researchers discovered connection characteristics in these commercially available automobiles
that make it easy to exchange information with others, according to the findings. The sharing of
information makes it feasible to carry out vehicle maintenance and to implement traffic safety
measures. As per Schmittner and Macher (2019), the high network connection of autonomous
vehicles increases the likelihood of an attack. In addition, they point out that the associated
infrastructure may be targeted by hackers for a number of different reasons. As a result, there are
now additional dangers to automotive cybersecurity to contend with. A study by the University
of California at Berkeley found that electronic components account for over half the total cost of
manufacturing a car with 100 million lines of computer code.

The consequence is that vehicle software upgrades are still discouraged, despite the fact that the
computer industry is much more proficient than the automotive industry in this area. In order to
meet this criteria, a full cyber-security assessment must be conducted, and this thesis proposes a
threat model to help with that. In order to analyse the security of the AUTOSAR benchmark and
provide suggestions on how to enhance security, the threat modelling approach must first
examine the standard's security. Before commencing this operation, current threat modelling
approaches are adjusted and altered in order to apply to the linked car and the AUTOSAR
standards, and then the procedure is restarted. To maintain a high level of security in the
architecture, future versions of AUTOSAR may take use of the threat modelling approach that
has been created. Every single one of the security flaws has been identified, documented, and
classified. With this information, we will be able to determine which components of the
AUTOSAR architecture are weak areas that need to be addressed specifically, as well as secure
the programs that rely on AUTOSAR as a basis for their functioning (McCarthy et al, 2014).

One of the goals of this dissertation is to have a better knowledge of the security measures
included in the connected car's software architecture. In this research, we will look into
AUTOSAR (Automotive Open System Architecture), a system standard that is used by both
automakers and electronic equipment manufacturers. Because of this standard, automotive
components designs will be simplified, allowing enterprises to cooperate on establishing the
standard while competing for market share in terms of implementation. ECUs, or embedded
controllers, make use of the AUTOSAR operating system to offer a standardised computer
system that relieves software developers of the need to depend on the hardware manufacturer.
Because it is a de facto industry standard, it is vital that it be examined thoroughly from a
security perspective. An automobile's internal network is a vital component of the vehicle's
operation. This network connects all of the electronic control units (ECUs) in the vehicle, and
each ECU is responsible for a certain function. As a result of the fact that AUTOSAR requires no
knowledge of the intricacies of the ECU, firms may develop software far more rapidly when
using it (Ivanova, 2017).

With the advent of self-driving automobiles, the automotive industry is entering a new realm of
security. It has previously been shown that traditional computer security is impossible to
implement, with disastrous consequences on several previous instances. Robotic vehicle
cybersecurity seems to be a new step that raises system complexity and opens the chance to even
more catastrophic failures in a robotic environment. However, unlike assaults on computer
systems, which often result in financial losses, a small-scale cyberattack on an autonomous
vehicle has the potential to cause human deaths. Although automobile security is commonly
referred to as "new," it has been the subject of extensive study for many years. In the past,
automobiles were thought of as closed systems, with electrical or digital controls maintained
separate from the mechanical systems that controlled the vehicle's vital components (Sanguino et
al, 2020). Hackers have been unable to gain access to or control a car as a consequence of the
physical requirements for an attack. According to some researchers and OEMs, hacking are
either impossible or insufficiently vulnerable if they need a physical connection. Encryption is a
kind of car cybersecurity solution that is available. These approaches have been demonstrated to
be useless for the CAN protocol, owing to the high computational and data overheads involved
with them. Researchers have proposed alternative methodologies such as Intrusion Detection
Systems in order to alleviate the inadequacies of cryptographic-based systems. For example,
Hafeez et al. provide an intrusion detection system that operates at the physical layer. It is
necessary to first estimate the frequency range of each transmitter and then train a neural
network to use the frequency range as a signature in order to recognise which controller or
transmitter is being used.

Using AUTOSAR for automobile cybersecurity and its recommendations

A growing number of vehicle safety-relevant capabilities are being introduced as humans move
toward fully connected and autonomous driving. As a result, more complicated security
procedures and a high degree of protection for vehicle platforms are becoming more critical than
they have ever been before. OEMs will also progressively build new business opportunities
based on high interconnectivity in the future, which will need the development of security
measures. This provides a clear mandate for the ongoing development of AUTOSAR Adaptive,
which will allow it to incorporate security applications considerably more firmly than before.

Additionally, additional IT security features such as hardware security components and the
potential implementation of intrusion - detection solutions will need to be taken into account
when further developing AUTOSAR Adaptive, with an integrated automobiles security approach
serving as its guiding principle.
In this context, it's vital to remember that AUTOSAR, as middleware, offers services that may be
utilised by a broad range of applications. As a result, even while AUTOSAR capabilities may be
used in a variety of ways across the vehicle design, they do not necessarily target the precise use
case which is required to achieve a certain demand.

UN R155 requires that practically every mitigation measure be supported by one or more
modules. AUTOSAR supplies one or more components that can enable the corresponding
technology platform for almost every solution for reducing needed under UN R155 AUTOSAR
even has capabilities that may be utilised to fulfil the criteria to a significant degree for a
significant number of these mitigations. In a nutshell, AUTOSAR succeeds on its promise, and
its numerous security modules are well suited to providing all of the mitigations needed by the
UNECE standards in a single package. When it comes to achieving UNECE cybersecurity
standards as a precondition for type approval, the degree to what they contribute to achieving
these requirements – that is, whether products meet these requirements entirely or simply partly
– is more complicated. In other words, although AUTOSAR may undoubtedly contribute to a
solution, it does not, by itself, give a complete solution. Developing security ideas for all
mitigations is required in order to accomplish holistic vehicle security, and these
conceptualizations must take into account both AUTOSAR modules and extra vehicle safety
systems including such hardware encryption modules (HSMs).

When used in this manner, AUTOSAR may serve as a valuable anchor and a powerful lever. For
the reason that Autosar now offers an extensive portfolio of cybersecurity products and services
that address, to varied degrees, the particular mitigations which must be taken into consideration
in accordance with UN Security Council Resolution 155. Thus, specifically taking into account
AUTOSAR security modules at relevant locations within the CSMS and including them in the
deployment of UNECE-compliant automotive cybersecurity makes sense. In order to create an
E/E infrastructure that is secured enough for type approval, manufacturers and distributors may
save time and money by effectively using the inherent capabilities of current AUTOSAR
security features to meet the risks and mitigations outlined in the UNECE standards. As a result,
AUTOSAR becomes a critical aspect in ensuring that automobiles have enough protection in
order to receive clearance. Overall automotive security, as defined by UNECE WP.29, will
eventually can go beyond AUTOSAR – from the cryptographic module (HSM) in a
microcontroller and the vehicle's private network to the OEM server side or vehicle security
operations centre (V-SOC), and throughout the entire supply chain and vehicle life cycle.
References

Acharya, S., Dvorkin, Y., Pandžić, H. and Karri, R., 2020. Cybersecurity of smart electric
vehicle charging: A power grid perspective. IEEE Access, 8, pp.214434-214453.

Amin, M. and Tariq, Z., 2015. Securing the car: How intrusive manufacturer-supplier approaches
can reduce cybersecurity vulnerabilities. Technology Innovation Management
Review, 5(1).

Bergin, D.L., 2015. Cyber-attack and defense simulation framework. The Journal of Defense
Modeling and Simulation, 12(4), pp.383-392.

Checkoway, S., McCoy, D., Kantor, B., Anderson, D., Shacham, H., Savage, S., Koscher, K.,
Czeskis, A., Roesner, F. and Kohno, T., 2011, August. Comprehensive experimental
analyses of automotive attack surfaces. In USENIX Security Symposium (Vol. 4, No.
447-462, p. 2021).

Cui, L., Hu, J., Park, B.B. and Bujanovic, P., 2018. Development of a simulation platform for
safety impact analysis considering vehicle dynamics, sensor errors, and communication
latencies: Assessing cooperative adaptive cruise control under cyber
attack. Transportation research part C: emerging technologies, 97, pp.1-22.

Eiza, M.H. and Ni, Q., 2017. Driving with sharks: Rethinking connected vehicles with vehicle
cybersecurity. IEEE Vehicular Technology Magazine, 12(2), pp.45-51.

Ivanova, Y., 2017. Modelling the impact of cyber attacks on the traffic control centre of an urban
automobile transport system by means of enhanced cybersecurity. In MATEC Web of
Conferences (Vol. 133, p. 07001). EDP Sciences.

Jadhav, A., 2021. Automotive Cybersecurity. In Automotive Embedded Systems (pp. 101-114).

Springer, Cham.

Jadhav, A., 2021. Automotive Cybersecurity. In Automotive Embedded Systems (pp. 101-114).

Springer, Cham.
Linkov, V., Zámečník, P., Havlíčková, D. and Pai, C.W., 2019. Human factors in the
cybersecurity of autonomous vehicles: Trends in current research. Frontiers in
psychology, 10, p.995.

Macher, G., Schmittner, C., Veledar, O. and Brenner, E., 2020, September. ISO/SAE DIS 21434
Automotive Cybersecurity Standard-In a Nutshell. In International Conference on
Computer Safety, Reliability, and Security (pp. 123-135). Springer, Cham.

Mäntymäki, M., 2016. ANALYSIS OF AUTOMOTIVE CYBERSECURITY.

McCarthy, C., Harnett, K. and Carter, A., 2014. A summary of cybersecurity best practices (No.
DOT HS 812 075). United States. National Highway Traffic Safety Administration.

Miller, C. and Valasek, C., 2015. Remote exploitation of an unaltered passenger vehicle. Black
Hat USA, 2015(S 91).

Obaid, M., Szalay, Z. and Török, Á., 2020. Reconsidering the cybersecurity framework in the
road transportation domain. Acta Polytech. Hung, 17, pp.57-83.

Ring, M., Frkat, D. and Schmiedecker, M., 2018. Cybersecurity evaluation of automotive e/e
architectures. In ACM Computer Science In Cars Symposium (CSCS 2018).

Sabatier, J., Merveillaut, M., Francisco, J.M., Guillemard, F. and Porcelatto, D., 2014. Lithium-
ion batteries modeling involving fractional differentiation. Journal of power sources, 262,
pp.36-43.

Sanguino, T.D.J.M., Domínguez, J.M.L. and de Carvalho Baptista, P., 2020. Cybersecurity
certification and auditing of automotive industry. In Advances in Transport Policy and
Planning (Vol. 5, pp. 95-124). Academic Press.

Saunders, M., Lewis, P. and Thornhill, A., 2003. Research methods forbusiness students. Essex:
Prentice Hall: Financial Times.

Saunders, M., Lewis, P.H.I.L.I.P. and Thornhill, A.D.R.I.A.N., 2007. Research

methods. Business Students 4th edition Pearson Education Limited, England.

Shahabadkar, P., Hebbal, S.S. and Prashant, S., 2012. Deployment of Interpretive Structural
Modeling methodology in supply chain management-an overview.
Sim, J., Saunders, B., Waterfield, J. and Kingstone, T., 2018. Can sample size in qualitative
research be determined a priori?. International Journal of Social Research
Methodology, 21(5), pp.619-634.

Villasenor, J.D., 2011. Ensuring hardware cybersecurity. Center for Technology Innovation at

Brookings.

Meisel, J., 2006. An analytic foundation for the Toyota Prius THS-II powertrain with a
comparison to a strong parallel hybrid-electric powertrain (No. 2006-01-0666). SAE
Technical Paper.

Gottehrer, G., 2015. Connected discovery: What the ubiquity of digital evidence means for
lawyers and litigation. Rich. JL & Tech., 22, p.1.

Farris, P., Pfeifer, P.E., Johnson, R.R. and McCormack, W., 2003. General Motors
OnStar. Darden Case No. UVA-M-0659.

Fitzgerald, M., 2016. General Motors relies on IoT to anticipate customers' needs. MIT Sloan
Management Review, 57(4).

Endsley, M.R., 2017. Autonomous driving systems: A preliminary naturalistic study of the Tesla
Model S. Journal of Cognitive Engineering and Decision Making, 11(3), pp.225-238.

Rigoll, G., 2007. The connecteddrive context server–flexible software architecture for a context
aware vehicle. Advanced Microsystems for Automotive Applications 2007, 201.

Heinecke, H., Schnelle, K.P., Fennel, H., Bortolazzi, J., Lundh, L., Leflour, J., Maté, J.L.,
Nishikawa, K. and Scharnhorst, T., 2004. Automotive open system architecture-an
industry-wide initiative to manage the complexity of emerging automotive e/e-
architectures (No. 2004-21-0042). SAE Technical Paper.

Fürst, S., Mössinger, J., Bunzel, S., Weber, T., Kirschke-Biller, F., Heitkämper, P., Kinkelin, G.,
Nishikawa, K. and Lange, K., 2009, October. AUTOSAR–A Worldwide Standard is on
the Road. In 14th International VDI Congress Electronic Systems for Vehicles, Baden-
Baden (Vol. 62, p. 5).
Fennel, H., Bunzel, S., Heinecke, H., Bielefeld, J., Fürst, S., Schnelle, K.P., Grote, W.,
Maldener, N., Weber, T., Wohlgemuth, F. and Ruh, J., 2006. Achievements and
exploitation of the AUTOSAR development partnership. Convergence, 2006, p.10.

Clark, G.W., Doran, M.V. and Andel, T.R., 2017, March. Cybersecurity issues in robotics.
In 2017 IEEE conference on cognitive and computational aspects of situation
management (CogSIMA) (pp. 1-5). IEEE.

Straub, J., 2017, June. An approach to detecting deliberately introduced defects and micro-
defects in 3D printed objects. In Pattern Recognition and Tracking XXVIII (Vol. 10203,
p. 102030L). International Society for Optics and Photonics.

Al Mamun, A., Al Mamun, M.A. and Shikfa, A., 2018, July. Challenges and mitigation of cyber
threat in automated vehicle: An integrated approach. In 2018 International Conference of
Electrical and Electronic Technologies for Automotive (pp. 1-6). IEEE.

Schmittner, C. and Macher, G., 2019, September. Automotive cybersecurity standards-relation

and overview. In International Conference on Computer Safety, Reliability, and
Security (pp. 153-165). Springer, Cham.

Scalas, M. and Giacinto, G., 2019, October. Automotive cybersecurity: Foundations for next-
generation vehicles. In 2019 2nd International Conference on new Trends in Computing
Sciences (ICTCS) (pp. 1-6). IEEE.

Schoitsch, E. and Schmittner, C., 2020, May. Ongoing Cybersecurity and Safety Standardization
Activities Related to Highly Automated/Autonomous Vehicles. In International Forum on
Advanced Microsystems for Automotive Applications (pp. 72-86). Springer, Cham.

Shan, H., He, K., Wang, B. and Fang, X., 2020, August. Road vehicles Cybersecurity system
evaluation method. In Journal of Physics: Conference Series (Vol. 1607, No. 1, p.
012054). IOP Publishing.