Trust Modelling for Security of IoT Devices

International Journal of Trend in Scientific Research and Development (IJTSRD)
Volume 4 Issue 4, June 2020 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470
@ IJTSRD | Unique Paper ID – IJTSRD31573 | Volume – 4 | Issue – 4 | May-June 2020 Page 1479
Trust Modelling for Security of IoT Devices
Karthik MV
Department of Computer Science & Amp; IT,
Jain Deemed-To-Be-university, Bengaluru, Karnataka, India
ABSTRACT
IoT (Internet of Things), represents many kinds of devices in the field,
connected to data-centres via various networks, submitting data, and allow
themselves to be controlled. Connected cameras, TV, media players, access
control systems, and wireless sensors are becoming pervasive. Their
applications include Retail Solutions, Home, Transportation and Automotive,
Industrial and Energy etc. This growth also represents security threat, as
several hacker attacks been launched using these devices as agents. We
explore the current environment and propose a quantitative and qualitative
trust model, using a multi-dimensional exploration space, based on the
hardware and software stack. This can be extended to any combination of IoT
devices, and dynamically updated as the type of applications, deployment
environment or any ingredients change.
KEYWORDS: Edge Computing, Security, Adaptive learning, Trust model, Threats,
Cloud Computing, Information Security
How to cite this paper:Karthik MV"Trust
Modelling for Security of IoT Devices"
Published in
International Journal
of Trend in Scientific
Research and
Development
(ijtsrd), ISSN: 2456-
6470, Volume-4 |
Issue-4, June 2020,
pp.1479-1485, URL:
www.ijtsrd.com/papers/ijtsrd31573.pdf
Copyright © 2020 by author(s) and
International Journal ofTrendinScientific
Research and Development Journal. This
is an Open Access article distributed
under the terms of
the Creative
CommonsAttribution
License (CC BY 4.0)
(http://creativecommons.org/licenses/by
/4.0)
1. INTRODUCTION:
Security concerns [1] abound with the emergence of IoT
devices in Cloud Computing. A recent DDOS (Distributed
Denial of Service) attack was launched using hijacked home
security cameras, while in another instance private video
clips were stolen and posted on Internet. Vulnerabilities in
other unprotected devices, such as home appliances (TV,
Fridge) on a network can be used to launch a cyber-attack.
IoT devices are constantly collecting data about an
environment or individuals, which can be potentiallyshared
with third parties compromising privacy. It can range from
personal preferences of web-browsing habits, TV channels
selection, or images from home security cameras. In
addition, there are security concerns if access controls to
these IoT devices are compromised. An example is of
someone hacking into a home control system toopengarage
doors or alter air-conditioning settings. While thelatter may
represent a minor inconvenience for a homeowner, if done
for many homes at once can result in an overload of thelocal
electric grid.
Furthermore, if these devices connect to a service provider
then its servers can be accessed via the devices to
compromise its security. If IoT devices are located in a
factory then an unauthorized access can be used toharmthe
equipment or products being manufactured. If these IoT
devices are deployed in a hospital, then patient care can be
compromised. At an individual level, it may mean incorrect
readings from a blood sugar monitor resulting in
inappropriate dosage of insulin, potentially with fatal
consequences.
2. BACKGROUND
Another emerging trend is a Cloud driven by things vs.
current Cloud Computing mostly driven by people, as
cameras and wireless sensors are becoming pervasive [2].
Growth and distribution between consumers and business
are shown in Figure 1. Their applications include Retail
Solutions, Home, Transportation and Automotive,Industrial
and Energy etc. An example of retail industry is Amazon’s
user and transact them. An example of Transportation and
Automotive is a Software Defined Cockpit in a commercial
aircraft, or an autonomous vehicle. An example of
manufacturing is a smart factory with robots or energy sav
health, print imaging, gaming and education are being
digitized at an unprecedented rate. The phrase “Internet of
things” was first used by British technology visionary Kevin
Aston in 1999. His perception was to think of “objects in
physical world connected by sensors”. Internet Architecture
Board (IAB) RFC 7452 provides the definition of IoT, as
follows: "Internet of Things" (IoT) denotes a trend where a
large number of embedded devices employ communication
services offered by Internet protocols.Manyofthesedevices,
often called "smart objects,’’ are not directly operated by
humans, but exist as components in buildings or vehicles, or
are spread out in the environment. Four basic
communication
1. Device to device
2. Device to cloud
3. Device to gateway
4. Backend data sharing model
We are more interested in #2 and #4, as both involve Cloud
services. An example is shown in figure 2, of home
IJTSRD31573

International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470
appliances such as a thermostat-controlled energy
management [3] Computer Science & Information
Technology (CS & IT) Another emerging trend is a Cloud
driven by things vs. current Cloud Computing mostly driven
by people, as cameras and wireless sensors are becoming
pervasive [2].
Growth and distribution between consumers and business
are shown in Figure 1. Their applications include Retail
Solutions, Home, Transportation and Automotive,Industrial
and Energy etc. An example of retail industry is Amazon’s
user-facing portalswherecustomerscanvisualizethingsand
transact them. An example of Transportation and
manufacturing is a smart factory with robots or energy
savings in a building.
Lastly, additional market segments such as health, print
imaging, gaming and education are being digitized at an
unprecedented rate. The phrase“Internetofthings”wasfirst
used by British technology visionary Kevin Aston in 1999.
perception was to think of “objects in physical world
connected by sensors”. Internet Architecture Board (IAB)
RFC 7452 provides the definition of IoT,asfollows:"Internet
of Things" (IoT) denotes a trend where a large number of
embedded devices employ communication services offered
by Internet protocols.
Many of these devices, often called "smart objects,’’ are not
directly operated by humans, but exist as components in
buildings or vehicles, or are spread out in the environment.
Four basic communication models for IoT are: Backenddata
sharing model We are more interested in #2 and #4, as both
involve Cloud services. Home appliances such as a
thermostat-controlled A/C connected to Cloud for better
Another emerging trend is a Cloud driven by things vs.
current Cloud Computing mostly driven by people, as
cameras and wireless sensors are becoming pervasive [2].
Growth of IOT devices, and distribution betweenconsumers
and business. Their applications include Retail Solutions,
Home, Transportation and Automotive, Industrial and
Energy etc. Ans where customers can visualize things and
transact them. An example of Transportation and
manufacturing is a smart-tings in a building. Lastly,
additional market segments such as health, print imaging,
gaming and education are being digitized at an
unprecedented rate. The phrase“Internetofthings”wasfirst
used by British technology visionary Kevin Aston in 1999.
perception was to think of “objects in physical world
connected by sensors”. Internet Architecture Board (IAB)
RFC 7452 provides the definition of IoT,asfollows:"Internet
of Things" (IoT) denotes a trend where a large number of
embedded devices employ communication services offered
by Internet protocols. Many of these devices, often called
"smart objects,’’ are not directly operated by humans, but
exist as components in buildings or vehicles, on models for
IoT are: We are more interested in #2 and #4, as both
involve Cloud services. An example is shown in A/C
connected to Cloud for better
3. SECURITY ATTACKS USING IOT DEVICES
For ensuring trust in IoT based Cloud Computing, it has to
start with a trusted environment, trusted protocols and
tamper proof components. Vendors need to provide “anti-
tamper” solutions. Software upgrades inthefieldareneeded
for any bug fixes during the lifetime of an IoT device. A
secure channel must exist to provide signed data that are
transmitted and installed in the field, e.g., on a car or TV at
home. In our door example, the vendor needs to provide an
anti-tamper solution, to prevent someone locally changing
the firmware or settings in an unauthorized manner.
Even remote software upgrades are authenticated, as
unprotected home appliances can be used to launch cyber-
attacks, e.g., someone using a collection of botnets to launch
a DDOS attack on a Cloud server, where a botnet refers to
one or more IoT devices being remotely controlled like a
robotic army. Besidessecurity,thereareprivacyconcerns,as
home sensors are collecting data about individuals that can
be shared with third parties for commercial and political
purposes.
Undesirable consequence may emerge if a third party can
remotely gain control, e.g., of a selfdriven car causing an
accident on the road, or someone with malice can access the
medicine dripmeters in a hospital with fatal consequences
for the patients. This can be avoided with a balanced
approach to interoperability and access control. This needs
to be addressed at different layers of architecture, and
within the protocol stacks between the devices.
Standardization and adoption of communication protocols
should specify when it is optimal to have standards.
Some vendors like to create a proprietary ecosystem of
compatible IoT products. This creates user lock-in to their
particular ecosystem, which from a vendor’s point of view is
desirable because a closed ecosystem approach can offer
benefits of security and reduces costs. However, from a
user’s point of view, such practices can create
interoperability problems with solutions from other
vendors, thereby limiting user’s choices in case of upgrades
or future system expansion.
As the frontiers of Cloud computing are expanding beyond
the walls of a datacenter to the extremes of a network,a new
term called Edge Computing is emerging. Itreferstothedata
analytics occurring at the sources of data generation. This is
bringing forth both new and existing security challenges,
Following classificationsdescribethetypesofsecurityissues
as related to the Edge Computing, with a few examples:
1. Identity authentication: By definition, the number of
players in Edge Computing is large and these may not
belong to the same organization. It is infeasible to verify
their 160 Computer Science & Information Technology
(CS & IT) identity in a foolproof manner. Trust needs to
be extended, as new customers buy their devices, such
as security cameras, and bring these online with a
remote registration.Central authoritythenmustdepend
on the ability of these remote customers to protecttheir
own devices.
2. Unauthorized access: Depending on the nature of
devices at the Edge, their access into data-centermaybe
bi-directional in nature. If someone hacks into a trusted
remote device, and retrieves its authentication
certificates to configure their owndevices,thenit will be

nearly impossible to differentiate between genuine or
fake users. Similarly, someone pretending to act as a
central computer can access the remote devices and get
critical user-data, such as on remote medical devices.
3. Denial of service attacks: An attack launched by
hijacking multiple remote devices and simultaneously
contacting the central server. This will cause the server
to be overloaded, denying access to genuine users in a
timely manner.
4. Data theft: Depending on where data is stored and for
how long opens the possibility of it being stolen. An
example is a security camera at home with local storage.
In event of a theft, it may be possible for an intruder to
simply remove the local storage, thus circumventingthe
purpose of a security camera. However, if camera
immediately uploads animagetoCloudupondetectinga
motion, then any physical tampering will not alter the
images of intruders.
5. Data integrity andfalsification:Akeydifferencebetween
confidentiality and integrity is that in the latter case, an
attacker doesn’t need to read the protected data, but
merely modify it, e.g., with a buffer overflow, rendering
it useless. This system level attack can happen if
multiple devices from different sourcesarewritingback
to a central server database.
6. Invasion of privacy: Since multipleplayersmaycombine
their data inputs from different sources to arrive at a
desired conclusion, e.g., for real-time traffic updates,
their identities need to be protected. This may include
an individual’s location, movements and any other
aspects of personal nature.
7. Activity monitoring: A cell phone that constantly pings
the signal tower, is sufficientforsomeonetomonitorthe
location of aphone’s owner, their movements etc.
Furthermore, if a remote app can turn on the
microphone or camera in a phone, then additional
information and activities can be monitored in anillegal
manner. Similar effects can be achieved with fixed
cameras at commercial or public locations, e.g., in a
shopping center.
8. Rooting of devices: Additional software can be installed
in the IoT devices without users’ permission. The
software can ‘root’ the device preventing detection and
have full access. There is no universal virus or malware
scanner for IoT. Some devices can be programmed to
selectively transmit data to a cloud service for
processing, e.g., a security camera which has a buffer of
15 seconds, but records and transmits a 30 seconds of
clip only if any motion is detected, for 15 secondsbefore
and 15 seconds after the motion is detected. This
reduces storage requirementsbutincreaseschancesofa
missed detection. Such devices are designed to render
service with minimal intervention, and yet they need to
be directed using voice activation or image recognition.
These and other devices can be used to conduct a DDOS
attack on the backend server, even in a serverless
architecture [4]. The attacker simply hijacks one or more
devices, and uses them to inundate the backend services.
This can be done by sending moredata,andmoreoften,from
the camera even when there is no motion detected.
4. SECURITY SOLUTIONS FOR IOT DEVICES
Solution level cost considerations involve technical factors
such as limited internal processing, memory resources or
power consumption demands.Vendorstryto reducetheunit
cost of devices by minimizingpartsandproductdesigncosts.
It is more expensive to designinteroperabilityfeaturesintoa
product and test for compliance with a standards
specification. A non-interoperable device may lack in
standards and the documented best practices. It may limit
the potential use of IoT device, and absence of these
standards can result in deviant behavior by IoT devices. It is
recognized that traditional Trusted Compute Boundary
(TCB) expands withEdgeComputingtoincludedomainsthat
are physically outside the control ofremotedeviceorcentral
data-center owners. The best theycandoistomonitor/track
a threat, identify an attacker, launch a recovery and prevent
false positives. These steps are outlined below:
1. Monitor/track a Threat: This is possible by establishing
a normal usage pattern for the IoT device, an example is
a security camera at home, which uploads data
whenever any motion is detected, e.g., wheneverpeople
go in and out. If the regular pattern for a home is no
more than a couple of dozen data uploads during a day,
then hundreds of data loads to the central server within
a few minutes may indicate that the device has been
compromised. It could be an attempt to cause a DOS
attack.
2. Identifying attackers: Once a threat is detected, then
attackers need to be identified. These could take the
form of an IP address of the IoT that is repeatedly
pinging the central server, to launch a denial of service
attack.
3. Attack recovery: This can take the form of blocking the
offending IP address. However, an attacker can corrupt
the critical data before the attacker’s presence is
detected. In such a case, frequent checkpoints must be
taken to do a rollback to the known good state.
4. Accidental and unintentional failures confused with
security attacks: Any detection method suffers from the
risks of false positives, e.g., mistaken flagging ofgenuine
access as a potential threat. An example of this is a stock
market trading computer that detects unusual activity,
which is genuine yet may flag a false alarm. Similar
situation can happen with security alarms due to false
sensor activity data etc. This calls for a learning system
that becomes smarter over time.
5. Data Integrity Protection: We previously described a
system level attack if multiple devices from different
sources are writing back to a central server database.
This can be protected by assigning a virtual partition or
container to the data coming from each distinct source,
and checking the address range of each access to
prevent data integrity of other users onthesameserver.
Internet Engineering Task Force (IETF) has identified the
problem of Interoperability, as many suppliersbuild“walled
gardens” that limit users to interoperate with a curated
subset of component providers, applications and services.
Interoperability solutions between IoT devices and backend
systems can exist at different layers of the architecture, and
at different levels within protocol stack betweenthedevices.

Key is the standardization and adoption of protocols, which
should specify when and where it is optimal to use
standards. More work is needed to ensure interoperability
within the cost constraints for Edge Computing to become
pervasive.
There are other regulatory and policy issues at play, such as
device data being collected and stored in a Cloud may cross-
jurisdictional boundaries, raising liability issues if the data
leaks. This is especially important if data is of personal
nature, e.g., related to shopping patterns or patient health
records.
5. TRUST MODELS FOR IOT DEVICES
Attacks have been made exploiting a component level
vulnerability. Most security systems are designed using
Capability models. A capability model usually takes into
account how various services are utilized. For example, we
can start with a multi-dimensional structure, composed of:
1. Hardware: An ASIC or programmable microcontroller.
2. Operating System: Windows, Linux, Android etc.
3. Applications: nature of application, and its privilege
level.
4. Manner in which various components, services and
utilities are deployed:
a. e.g., kernel, library services, files accesses,
b. Manner in which objects (username, application,
function) get authenticated,
c. What kind of cryptography is utilized, e.g., strength of
MD5 vs. SHA256.
We propose to evaluate a given HW and SW solution
components composed of one or more IOT devices
connected to a Cloud server, based on the robustness and
trustworthiness of this entire solution stack, with a
multiplicative serialized model, e.g., in the following order:
1. Native compiled code is trusted more
2. Then anything using an external library
3. Lastly, any third party SW attempting to integrate
Using the above method, it is possible for us toevaluatetrust
of different operating systems with applications from
diverse fields. Goal is to create a framework for evaluating
and assigning a security score to each layer and then
compute a composite score. A given application can be
disassembled to see whether it is using a kernel service, or a
utility in the user-space, or a built-inlibrary etc.
For each component in the stack, a list of orthogonal
properties are established followed by an objective scoring
system for each property. Numerical score for a utility
function depends on the manner in which it is accessed, e.g.,
read (as a call by value), or a write (call by reference). A
Security Score can computed by answeringa setofquestions
by a user or automatically computed by a testing tool.
Example of questions include:
Whether a salt is used hash passwords?
Which algorithm is used for hashing: MD5 or SHA256?
Does the communication channel use SSL and which
version of TLS is being used?
What is the version of MYSQL in operation?
Another SecurityScoredeterminationmethod: Whetherport
3306 used by MySQL is open to the world or just to the
application servers that use the MySQL database. This score
can be continuously updated during the operations. More
importantly, it needs to be updated after a maintenance or
upgrade action is completed. Security Score questionnaire
may focus core calculation focusesonthesystemoperations.
An OS without the latest patch can be at a security risk.
Security Score computations has two outputs:
1. Probability of a successful attack:
will succeed?
2. Probable Impact of a successful attack:
succeeds?
The Security Score (S) can be computed as follows:
S = 1 - Pa * Pi
Where:
Pa - Probability of the attack in the range 0 to 1
Pi - Probable impact if the attack succeeds in the range 0to1
Pa * Pi - is the expected loss
This score is for a single component. By describing the
security different components and their computed.
The factors that affect the probability of attack include:
Presence of a vulnerability existing and known to
attackers
Level of focus on products of this type by hackers
History of exploitation of this product type
Probable impact of a security attack is defined as the sum of
any regulatory fines, reputational damage and operational
loss. This represents the resulting loss of trust in product
and services. This needs constant monitoring for security
breaches and policy updating [5]. The first step in the
modelling is describe the whole system in terms of its
components hierarchically organized and security or
parallel.
Another SecurityScoredeterminationmethod: Whetherport
3306 used by MySQL is open to the just to the application
servers that use the MySQL database. This score can be
continuously updated during the operations. More
importantly, it needs to be updated after a maintenance or
upgrade action is completed. Security Score questionnaire
may focus on the best practices during development.
Automated score calculation focuses on the system
operations. An OS without the latest patch can be at a
Security Score computations has two outputs: Probabilityof
a successful attack: What is the probability that an attack on
this device Probable Impact of a successful attack: What is
the probable impact if the attack The Security Score (S) can
be computed as follows: of the attack in the range 0 to 1
Probable impact if the attack succeeds in the range 0 to 1 is
the expected loss This score is for a single component. By
describing the security-wise relationship among the
different components and their individual security scores,
the whole system security score can be.
The factors that affect the probability of attack include:
Presence of a vulnerability existing and known to attackers
Level of focus on products of this type by hackers of
exploitation of this product type Probable impact of a

security attack is defined as the sum of any regulatory fines,
reputational damage and operational loss. This represents
the resulting loss of trust in product and services. Nitoring
for security breaches and policy updating [5].
The first step in the modelling is describe the whole system
in terms of its components hierarchically organized and
security-wise connectionsbetweenthecomponentscouldbe
serial The direction lines in figure 3 represent the security-
wise relationship between the blocks in a system. In figure
3(a), to all the blocks should be secure for system to be
secure and provide the required functionality.Infigure3(b),
any one of the blocks should be secure for the system to
provide the required functionality.
The composite Security Score can be computed by applying
parallel reliability rules [6]. Another Security Score
determinationmethod: Whetherport3306used byMySQL is
open to the just to the application servers that use the
MySQL database. This score can be continuously updated
during the operations. More importantly, it needs to be
updated after as on the best practices during development.
Automated score calculation focuses on the system
operations. An OS without the latest patch can be at a What
is the probability that an attack on this device What is the
probable impact if the attack wise relationship among the
individual security scores, the whole system security score
can be Probable impact of a security attack is defined as the
sum of any regulatory fines, reputational damage and
operational loss. This represents the resultinglossoftrustin
product and services. The first step in the modelling is
describe the whole system in terms of its components wise
connections between the components could be serial wise
relationship between the blocksina system.Infigure3(a),to
all the blocks should be secure for system to be secure and
provide the should be secure for the system to can be
comuted by applying
6. AN EXAMPLE STUDY
Raspberry PI is the de facto choice and starting point for
many IoT devices. This choice is driven by its ubiquity and
low price, making it a popular controller for many homeand
entry level appliances. An higher installed basealsomakesit
an attractive target for hackers, therefore weevaluateditfor
our IoT trust model. For our sample system, we restricted
probability values to High (0.9),Medium(0.6)andLow(0.3).
Similarly, the impact Medium (0.6) and Low (0.3). We took
an implementation of a Raspberry Pi Model 3B with
Raspbian OS Ver 4.14 released on 2018-4-18 as a reference
system for trust scoring [7]. The base Raspberry Pi system
comes with a microSD card, which holds the OS and can be
used to install additional software. The factory settings and
factory shipped software packages for the OS were used for
trust scoring. No packages were updated. Once the basic
model trust scoring was complete, we proceed complete the
Raspberry Pi based Security Camera setup [8]. Following
additional software components were installed.
1. MongoDB
2. Rabbit MQ
3. AWS IOT client
4. MotionPie software
We use Mongo DB to have a NAS(Network Attachedstorage)
of images, and the AWS IoT client to connect with Amazon’s
backend service for cloud storage. MotionPie is an image
processing software to detect motion,andthendecide which
video clips discarded.
STUDY OF TRUST SCORING
Raspberry PI is the de facto choice and starting point for
many IoT devices. This choice is driven by its ubiquity and
low price, making it a popular controller for many homeand
entry level appliances. An higher installed basealsomakesit
an attractive target for hackers, therefore weevaluateditfor
our IoT trust model. For our sample system, we restricted
probability values to High (0.9),Medium(0.6)andLow(0.3).
Similarly, the impact values were also High (0.9),Wetook an
implementation of a Raspberry Pi Model 3B with Raspbian
OS Ver 4.14 released on 18 as a reference system for trust
scoring [7]. The base Raspberry Pisystemcomes witha card,
which holds the OS and can be used to install additional
software. The factory settings and factory shipped software
packages for the OS were used for trust scoring.Nopackages
were updated. Once the basic model trust scoring was
complete, we proceed complete the Raspberry Pi based
Security Camera setup [8]. Following additional software
components were installed.
We use Mongo DB to have a NAS(Network Attachedstorage)
of images, and the AWS IoT client to connect with Amazon’s
backend service for cloud storage. MotionPie is an image
processing software to detect motion,andthendecide which
video clips need to be saved or Raspberry PI is the de facto
choice and starting point for many IoT devices.Thischoiceis
driven by its ubiquity and low price, making it a popular
controller for many home and entry level appliances. An
higher installed base also makes it an attractive target for
hackers, therefore we evaluated it for our IoT trust model.
For our sample system, we restricted probability values to
values were also High (0.9).
We took an implementation of a Raspberry PiModel 3Bwith
Raspbian OS Ver 4.14 released on18 as a reference system
for trust scoring [7]. The base Raspberry Pi system comes
with a card, which holds the OS and can be used to install
additional software. The factorysettingsandfactoryshipped
software packages for the OS were used for trust scoring. No
packages were updated. Once the basic model trust scoring
was complete, we proceeded to complete the Raspberry Pi
based Security Camera setup [8]. Following additional
software We use Mongo DB to have a NAS (Network
Attached storage) of images, and the AWS IoT client to
connect with Amazon’s backend service for cloud storage.
MotionPie is an image need to be saved
A problem with this security camera prototype is that
someone with a physical access to local system can easily
switch the software. There is no method to check if the
system software is authenticated at boot time, so the base
hardware setup has a high pro The impact probability of
such attack is also high (0.9) as the base system can be fully
compromised. In the default setup,theuser nameis“admin”,
and password is blank. It is easy for someone to remotely
hijack and use this camera in a Mirai botnet attack [9]. After
the password has been changed, and if the camera is
installed behind a secure firewall, the probability of such an
attack is medium (0.6). However, the impact probability is
high (0.9). Our proposed system us IOT security model [10],
with X.509 certification with asymmetric keys [8]. On the
backend, where the images are stored, the securityishigh so

probability of an attack is low (0.3) and impact probabilityis
also low (0.3), since the images have a based storage. Even
though the local system can be attacked with a higher
probability (0.9) and medium impact (0.6) Overall, we have
the Raspberry hardware and software components in series
security itself is in series with two parallel storage systems
security
what we have so far:
Sc = 1 - 0.9 * 0.9 = 1
Ss = 1 - 0.6 * 0.9 = 1
and for the storage systems,
Sgc= 1 - 0.3 * 0.3 = 0.91
Sgl= 1 - 0.9
Where, Sc is the security of camera, Ssis the security of
software, Sgc is the security of cloud storage, and Sgl is the
security of local storage. As cloud storage and local storage
are in parallel providing redundant functionality, the
secured chaining rule:
system software is authenticated at boot time, so the base
hardware setup has a high probability (0.9) of an attack.The
impact probability of such attack is also high (0.9) as the
base system can be fully In the default setup, the user name
is “admin”, and password is blank. It is easy for someone to
this camera in a Mirai botnet attack [9]. After the password
has been changed, and if the camera is installed behind a
secure firewall, the probability of such an attack is medium
(0.6). However, the impact probability is high (0.9). Our
proposed system us IOT security model [10], with X.509
certification with asymmetric keys [8]. On the backend,
where the images are stored, the security is high so
probability of an attack is low (0.3) and impact probabilityis
also low (0.3), since the images have a local storage as we as
well cloud-based storage. Even though the local system can
be attacked with a higher probability (0.9) and Overall, we
have the Raspberry hardware and software components in
series security in series with two parallel storage systems
security-wise.
system software is ability (0.9) of an attack. The impact
probability of such attack is also high (0.9) as the base
system can be fully In the default setup, the user name is
“admin”, and password is blank. It iseasyforsomeonetothis
camera in a Mirai botnet attack [9]. After the password has
been changed, and if the camera is installed behind a secure
firewall, the probability of such an attack is medium (0.6).
However, the impact probability is high (0.9). Our proposed
system uses AWS IOT security model [10], with X.509
certification with asymmetric keys [8]. On the backend,
where the images are stored, the security is high so
probability of an attack is low (0.3) and local storage as we
as well cloud-based storage. Even though the local system
can be attacked with a higher probability (0.9) and Overall,
we have the Raspberry hardware and software components
in series security-wise, which wise. Atcomponentlevel,here
is Where, Sc is the security of camera, Ssis the security of
software, Sgc is the security of cloud storage, and Sgl is the
security of local storage. As cloud storage and local storage
are in parallel rity score can be computed using reliability
parallel.
Finally, the end-to-endsystemlevel security protectionscore
for an attack is a composite of three scores = 0.19 * 0.46 *
0.9514 = 0.07which is only 7% or very low. This means that
the entire camera system is prone to attacks. However, we
can still use it due to our added security measures of a
strengthened password, dual storage in the local and cloud
based databases etc. Thus, one of the two paths needs to be
secured to continue the required functionality: Path (a)
Camera → Software → Local NAS, or Path (b) Camera →
Software → Cloud Storage. Note that all past images will still
be preserved even if the system is compromised up to the
point of intrusion, e.g., if someone physically removes the
microSD card on a security camera. If a home or business
uses such a system, it may need multiple cameras soifone of
them is compromised, others will continue the surveillance.
An example is of 5 Pi cameras, with a shared local NAS and
common cloud storages.
The Security Score for the camera and software part is
computed as 1 - (1 - 0.19*.46)^5 =0.36. The entire system
security will be 0.36*0.9514 = 0.34 or 34%. This improves
the total system security by almost 5X. Another way to
achieve a better security is by making it harder to
compromise a single camera system, e.g., by putting it in a
cage so its microSD card can’t be easily replaced. Then the
probability of a physical attack goes from high to low, such
that Sc = 1 - 0.3 * 0.9 = 1 - 0.27 = 0.73. The overall score for
such a single camera system would be 0.73*0.46*0.9514 =
0.32, or 32%, which is almost same as our 5 parallel cameras
system, albeit at a much cheaper cost. However, it also
represents a single point of failure, so the real choice may be
a combination of both. This can be achieved by using two
secure camera systems in parallel, as redundancy is
important to improve security.
7. SUMMARY
In this paper we review the scope of various IoT (Internet of
Things) devices in the field that are bi-directionally
connected to data-centers (in-house or cloud) via various
networks. Then we look at the nature of security issues, and
mechanisms to quantify risk associated with the complete
hardware and software stack, with an example of a typical
surveillance camera system. We calculated system security,
and suggested ways to improve it. Our proposedmethodcan
be extended to evaluate any IoT system, and improve its
end-to-end security profile.
8. REFERENCES:
[1] N. K. Sehgal, S. Sohoni, Y. Xiong, D. Fritz, W. Mulia,andJ.
M. Acken, “A Cross Section of the Issues and Research
Activities Related to Both Information Security and
Cloud Computing,” IETETechnical Review, Volume 28,
Issue 4 [p. 279-291], 2011
[2] https://www.statista.com/statistics/370350/internet-
of-things-installed-base-by-category/
[3] N. K. Sehgal, PCP. Bhatt, “ Cloud Computing: Concepts
and Practices,”SpringerPublications©2018,ISBN 978-
3-319-77839-6

[4] https://medium.com/@MarutiTech/what-is-
serverless-architecture-what-are-its-criticisms-
anddrawbacks-928659f9899a
[5] R. Sandhu, A.S. Sohal and S. K. Sood, “Identification of
malicious Edge Devices in fog computing
Environments,” Information Security Journal: AGlobal
Perspective, Volume 26, 2017, Issue 5. Computer
Science & Information Technology (CS & IT) 167
[6] http://web4.uwindsor.ca/users/f/fbaki/85-
222.nsf/0/b7d85091e772a10185256f84007be5c1/$F
ILE/Lecture_07_ch6_222_w05_s5.pdf
[7] https://www.raspberrypi.org/downloads/raspbian/
[8] https://pimylifeup.com/raspberry-pi-security-
camera/
[9] https://www.sentinelone.com/blog/iot-botnet-
attacks-mirai-source-code/
[10] https://aws.amazon.com/blogs/iot/understanding-
the-aws-iot-security-model

Trust Modelling for Security of IoT Devices

More Related Content

What's hot (20)

Similar to Trust Modelling for Security of IoT Devices (20)

More from ijtsrd (20)

Recently uploaded (20)

Trust Modelling for Security of IoT Devices