Design of a Hybrid Authentication Technique for User and Device Authentication for an integrated IOT environment using Blockchain Authentication and Artificial Intelligence

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 12 | Dec 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 274
Design of a Hybrid Authentication Technique for User and Device
Authentication for an integrated IOT environment using Blockchain
Authentication and Artificial Intelligence
Ashish Dibouliya
Principal Architect – Data, Connecticut, USA
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Attacks against Internet of Things (IoT) devices
have increased dramatically in recent years. Attacks on vital
internet infrastructure systems have beencarriedoutbycyber
thieves using hacked IoT devices. Information security
researchers are under increasing pressure to provide new
tactics and solutions to protect susceptible devices at alllevels
following recent assaults. When it comes to home networks,
malevolent actors don't care about the smart devices that are
part of them. National and global governments are placing a
high value on securing critical IoT networks like the electrical
grid and water distribution systems. Many different
technologies are being applied to the problem of Internet of
Things security. These include blockchain, AI, and edge/fog
computing (IoT). Authentication and authorization are two
legs of the CIA's crucial triad of components. However, due to
the scale of IoT networks and the limited resources of devices,
standard authentication and authorization mechanisms are
unsuitable for handling security. We employ blockchain
technology as a lightweight and straightforward
authentication solution for IoT devices. The use of artificial
intelligence (AI) to aid in decision-making and carry out
predictive analytics in a wide variety of contexts is gaining
popularity. In recent years, blockchain (a peer-to-peer
distributed system) has also been used to enable AI
applications, such as secure data sharing (formodeltraining),
data privacy, trustworthy AI decision making, and
decentralized AI. The purpose of this effort is to create a
hybrid authentication technique based on Blockchain
authentication and AI for user and device authentication in a
fully interconnected IOT setting.
Keywords: IOT (Internet of Things), WSN (Wireless Sensor
Networks), Blockchain Authentication, Artificial Intelligence,
Hybrid Authentication
1. INTRODUCTION
The concept of the InternetofThings(IoT)israpidly
growing and gaining popularity. Many of our regular tasks
and routines would be impossible without the Internet of
Things. New IoT applications in areas including smart
homes, smart healthcare, smart cities, industry 4.0, wireless
sensor networks (WSN), and smart farming have the
potential to introduce and advance the concept of a smart
world. Security measures at several layers, along with the
capacity to analyze and regulateall data andinformation,are
essential for the platforms, networks, devices1, and
applications that make up the Internet of Things (IoT)
ecosystem. The devices can be easily compromised, leaving
users open to identity theft and meddling with their data.
Figure: 1 Basic Blockchain Structure
The ledger must be generated and maintained by each node
in the blockchain ecosystem. To participate, a participant
must be able to read and write transactions. There are two
categories of participants: read-only and read-write. To put
it another way, miners can add transactions to the chain as
unique nodes. An established technology, blockchainholdsa
lot of promise. It is regarded as a fundamental approach in a
decentralized network environment since it is distributed,
decentralized, andunchanging.Fortheblockchainnetwork's
peers to accept a block, miners must complete proof of work
(PoW). Figure 1 depicts the blockchain's basic structure.For
securing the Internet of Things (IoT), Blockchain has been
hailed as an ideal option. To ensure the safety of the IoT
ecosystem, we have also benefited from this technology.Itis
our primary purpose in this research project to provide an
authentication system based on the Ethereum Blockchain,
which is a public blockchain and provides distributed and
decentralized security for IoT devices. Using this method,
users and their gadgets may communicate in a secure
environment. There are three sections in which the
document is broken down: 1st section includes introduction
part of the paper. Second, there is a section called Related
Work that provides an overview of the blockchain and
includes work done by other authors to secure IoT with
blockchain. Third is the ProposedTechnique, whichincludes
the results and discussion of the proposed method, and
finally Section V is the Conclusion.

1.1 Internet of things paradigm
IoT is now a reality that pervades many aspects of
our life, and it is only going to becoming more prevalent in
the future. Although IoT was formerly viewed as a science
fiction concept, it has now become a reality thanks to the
work of several researchers who see it as one of the most
significant technicalshiftsofthetwenty-firstcentury.Asingle
linked thing per person was achieved in 2008, and current
estimates place the number at roughly 26 smart objects per
person by 2020. There are still many challenges that need to
be overcome before the Internet of Things can be used to its
full potential. These challenges include but are not limited to
computing constraints imposed by Internet of Things (IoT)
devices; heterogeneity; identification; power supply; data
storage/processing;amongotherthings.Securityandprivacy
are two of the most pressing issues, considering the
widespread use of smart devicesineverydaylife.IoTsecurity
solutions aren't applicable in general because of the
restrictions of IoT components, which have minimal
capabilities in both energy and processing resources.
As a result, they cannot implement elaborate
schemes supporting security. Unfortunately, this makes
things even more difficult. Most prevalent IoT exploits, and
vulnerabilities have been documented by the OWASP
Internet of Things Project.
1.2 IoT and Machine learning algorithms
Most devices that participate on the internet of
things are autonomous and have modest power
requirements. These nodescancollectdata,actbasedonthat
data, and communicate with one another and with other
entities thanks to the use of sensors, actuators, and even the
Cloud. When the complexity of the algorithms became too
difficult to manage, attention turned from artificial
intelligence (AI) to machine learning (ML).Machinelearning
methods are used in numerous modern contexts, from
statistical analysis and prediction to practical tasks like
speech recognition and fraud detection. Algorithms and
methods used in machine learning have been impacted by
many disciplines, including mathematics, neuroscience,
statistics, and computer science. To test an algorithm's
ability to learn from data, it is common to split the
development of the algorithm into two distinct phases: the
training phase and the verification phase. Most today's
machine learning algorithms fall under the categories of
unsupervised, supervised, and reinforcement learning. For
the first-class training phase to be successful, a labelled data
set is required so that the studied relationships can be
represented. In contrast to the first type, the input/output
pairings for the unsupervised learning techniques are not
provided. The primary focus of this activity isonlocatingthe
links between the various data points. The third type, online
learning, describes situations in which an agent acquires
knowledge about how to behave by engaging in trial-and-
error interactions with an interactive environment.
1.3 Objective of Paper
Design and development of an authentication
technique which employees hybrid algorithms/techniques
such as Blockchain AuthenticationandArtificial Intelligence,
for seamless user as well as device authentication in an IOT
network.
2. LITERATURE SURVEY
Achraf Fayad et. All (2019) Numerous studies have been
done to propose novel methodsandsystemsfor ensuringthe
authentication of IoT devices. Many existing efforts do not
meet all of the IoT's objectives, such as permitting the
mobility of nodes or guaranteeing scalability through
decentralised techniques or making it easy to integrate new
devices and services. The authors of this paper put up a
simple way of authentication in this situation. Using a
blockchain, which is entirely decentralised, is at the heart of
our answer to this problem. As a result, it fulfils all of the
previously described conditions. Real-world
implementations of our method were used to test the
validity of our findings. It was evident from theresultsthatit
was low in weight. [1]
Aissam Outchkoucht et. All (2017) Even as the
mechanisms for actuation, communication, and control
improve and become more widespread, the risks they pose
to user trust and the Internet of Things as a whole, when not
adequately addressed, are also increasing in sophistication
and pervasiveness. This investigation has centered on the
IoT (Internet of Things). It proposed a framework with the
intention of fixing two issues: I those that arise from using a
centralized design, and (ii) the need to transmit the
administration of access control from a central body to the
nodes of the network. Indeed, the limited devices of the
Internet of Things do not have the computational resources
or data storage space required to support a decentralized
access control system. The sheernumber ofconnectedsmart
devices can be overwhelming in an IoT setting, posing a
number of regulatory issues related to regulating user
access. As a result, security officers and managers often
resort to enforcing "static policies," in which all security and
access control rules are written by a single person. [2]
Soumyashree S. Panda et. All (2019) The Internet of
Things (IoT) and its many uses have become an integral part
of our everyday lives. Without the need for human
intervention, a number of smart devices can interact with
one other. To participate in an autonomous system, only
genuine entities should be able to participate. As a result,
this study proposes an IoT system authentication scheme.
Blockchain is the decentralisedledgertechnologythatwould
be used in the proposed system, not a central authority. [3]

Khadija Fazal et. All (2020) Authenticated users must be
able to access and retrieve data from each device. Access to
these gadgets and the communications they provide should
be safe in most instances. IoT security is a key roadblock to
widespread adoption and implementationofthetechnology,
for a variety of reasons, including the fact that it isextremely
vulnerable to cyberattacks. Authors present a blockchain-
based Ethereum authentication system forconstructingsafe
zones in the internet of things. Various IoT settings,services,
and situations can benefit from the authors' proposed
strategy. We can be confident in the security of our system
because we are using a public form of Blockchain.To test the
suggested approach, we want to write a smart contract in
solidity using the Remix IDE and analyse its performance in
terms of cost and time. [4]
Dr. Sakthi Kumaresh et. All (2021) In this article, a
Decentralized Artificial Intelligence-enabled Blockchain
network is implemented using the confluence of AI and
Blockchain technology (DAIBCN). The proposed DAIBC
method uses a decision tree model to automatically identify
miner nodes. There is less requirement for miner nodes to
perform complex hash functions in order to be eligible for
reward. The testing findings reveal that the suggested AI-
enabled approach saves a significant amount of energy and
also reduces the time it takes to complete a transaction
because the node selection is done automatically. [5]
Praveen Kumar Kollu et. All (2021) Blockhain
development has spurred continuing study in a variety of
theoretical and functional areas. Decentralization, identity,
and trust are three of today's most pressing technological
issues, and the blockchain is widely seen as a promising
answer even as it is still in its infancy. When it comes to
finding the best way to store and access cloud data, the
groundbreaking blockchain is an invaluable resource. This
article examines the potential of blockchain technology to
defend cloud computing. A cloud computing data security
mechanism is also presented in this paper. This application
uses smart contracts and permission lists to secure data
security. [6]
Shanshan Zhao et. All (2019) Data is generated, collected,
processed, sent, and stored by several organisations in an
IIoT ecosystem. Blockchaining both IIoT entities and
business processes is highly sought after by the sectors. The
Internet of Things (IIoT) is predicted to be widely
implemented in sectors as a result of fast technological
advancements and new business models. In this article,
we've looked at blockchain and IIoT integration from an
industrial standpoint. Introduced is a blockchain-enabled
architecture for the Internet of Things (IoT). The most
important applications and issues are covered. Blockchain-
enabled IIoT research issues and potential trends were also
examined by the authors. [7]
Muhammad Tahir et. All (2020) Authentication, privacy,
chaos, mining, and administration are just a fewofthe issues
that the growing IoT market and the Blockchain ecosystem
are creating. In IoT applications, mutual authentication is
essential because it gives security to users and assures the
validity of data and protects their privacy. The mutual
authentication of IoT devices has been separated into
distinct classes based on the authentication and assessment
methodologies. A innovative approach for mutual
authentication and authorisation is presented in this work
by the authors. Distributed joint conditional probability and
the selection of random numbers are used to establish
robustness in the proposed framework. The performance
assessment and results analysis reveal that the suggested
framework achieves better security strength, fewer
communication overhead costs, and consumes less time in
processing data together with increased security. [8]
Dr Mohd Javaid et. All (2021) In addition to ensuring the
safety of all communications between the many
interconnected smart devices, the blockchain also keeps a
permanent record of every single message ever sent.
Individuals' identities could also be protectedvia blockchain
technology. Due to its deceptive nature, this technology
enables its users to establish a safe and dependable online
persona. A Blockchain identity might be used for anything
from a single click to accessing a whole suite of applications
and services or even forging digital signatures. Blockchain
could be useful for SMEs and suppliers as it offers a secure,
high-quality transactional data source. Furthermore, a shop
must guarantee that all of its productsareofuniformquality.
[9]
Bhabendu K. Mohanta et. All (2019) The Internet of
Things (or IoT) is a relatively new concept. Smart homes,
smart cities, smart transportation, and smart healthcare
systems are just a few examples of how IoT is enhancing
human well-being in a variety of contexts. Real-time
monitoring and control are available inall circumstances.To
test the efficacy of the DecAuth (Decentralized
Authentication) system, this paper proposes a Blockchain-
based ethereum platform implementation. An IoT device
may be authenticated in a decentralised manner, according
to the experimental results. Also, according to the study, the
suggested system is impenetrable by the currently known
assaults. [10]
Jesse Yli-Huumo et. All (2016) The Bitcoin cryptocurrency
is powered by blockchain technology. An open ledger of all
transactions is maintained in a decentralized setting that is
accessible to everyone. Blockchainaimstogiveitsusers with
complete anonymity, safety, privacy, and transparency.
However, these characteristics pose several technological
issues and constraints that must be addressed.Authorsused
the systematic mapping study approach to map all relevant
studies on Blockchain technology tobettergraspthepresent
state of the field. The study's main objective was to map out

the existing state and future directions of blockchain
technology. Aside from the technical viewpoint, writers in
this study did not offer viewpoints on economics, law,
business, or regulation. From scientific databases, 41
primary publications were retrieved and examined by the
authors. [11]
3. METHODOLOGY
3.1 Proposed Work Block Diagram
Figure: 3.1 Proposed Work Block Diagram
The proposed system is aimed for seamless
authentication of User(s)&Device(s)inIOT-WSN Network’s,
the system employs a block – chain of user/device
authentication entries, which is used to train a Function
Fitting Neural Network, which is used to ascertain the
authenticity of new block - chain entry for user/device
authentication, in form of a transaction authenticity score.
Apart from block - chain based login access;
secondary authentication is provided basedUserandDevice
parameters listed below. The user/device parameters are
used to train another pattern matching neural network,
which is used to provide User/Device Authentication Score.
Block-Chain based authentication score is combined with
User/Device authentication score using custom machine
intelligence to generate final authentication score, which is
used to provide multi-layer access to various services and
resources hosted by the IOT – WSN Network.
About LEVENBERG-MARQUARDT
To calculate the minimum of a function (linear or
nonlinear) over a space of parameters, the well-known
Levenberg-Marquardt algorithm (LMA)isoftenemployed.A
function, such as a quadratic, is used internally to mimic a
reliable portion of the goal function. Once a good match is
established, the trusted area grows to include the new
addition. The Levenberg-Marquardt methodology is
sensitive to the initial beginning parameters, as is the case
with many numerical algorithms. This article describes the
training and testing phases of the Levenberg-Marquardt
algorithm for learning artificial neural networks.
Why LEVENBERG-MARQUARDT:
The created LM-ANN model is more sensitive for
prediction of the inflows when the statistics ofthelong-term
and seasonal-term outputs arecompared.Further,thespeed
and reliability of theLM-ANN techniquemakeitapplicableto
the modelling of a wide range of hydrological factors. For a
given collection of training data, the Levenberg Marquardt
(LM) training function achieved the best performance (the
smallest MSE). Levenberg Marquardt is found to bethemost
effective predictive algorithm through extensive testing.
Key Benefits of LEVENBERG-MARQUARDT Algorithm:
1. The LM algorithm is more robust and has more stable
convergence than thee GN algorithm
2. This algorithm is mostly used for training small and
medium sized problems in ANN.
3. The advantage of the approach is that, once thenetwork
is trained, it allows instantaneous evaluation of
solutions at any desired number of points; thus,
spending little computing time.
4. Levenberg-Marquardt algorithm gives the best results
with high performance.
5. The Levenberg-Marquardt (LM) gradient descent
algorithm is used extensively forthetrainingofArtificial
Neural Networks (ANN).
6. The Levenberg–Marquardt algorithm provides a
numerical solution to the problem of minimizing a
(generally nonlinear) function.
7. The LEVENBERG-MARQUARDT method is a single-shot
method which attempts to find the local fit-statistic
minimum nearest to the starting point.
How LEVENBERG-MARQUARDT Helps
In the proposed work Levenberg–Marquardt
algorithm Used for train artificial neural network and
process some functions than some mathematical calculation

will be perform then train function, then plot error and
performance plot and calculated estimated distance with
respect or transmit power. if any node tries to establish
connection in our network, if this is in our range then our
access controller provided true key for establish connection
with some leniency otherwise if this is in outside the range
then access controller providedfakerandomkeyforprivacy.
3.1.1 User & Device Parameters Used
User Parameters
S.No Parameter Name Narration
1. User
Name/Password
Primary Access Control & User
Identification
2. Keyboard/Touch
Handling Profile
2nd Factor Authenticator Using
Comparison of Keyboard/Touch
Handling With Previously
Known History
3. Fingerprint Scan 2nd Factor Authenticator Using
Biometric – Fingerprint Scan
Using Minutiae Extraction
&Euclidian Distance
Computation
4. Signature Using
DSC
2nd Factor Authentication Using
Digital Signature Certificate
Provided By Trusted Authority
Such As CCA In India
Table: 3.1 User Parameters
Device Parameters
S.No Parameter Name Narration
1. MAC Address Primary Device Identifier, Using
Unique Machine Address Used
For Direct Node Address Usage
or Referenced Usage Such As IP
Address Based Redirection
2. Node
Configuration
Computing Configuration of The
Device Such As Processor Speed,
Number of Cores, RAM, Flash
ROM, Architecture Etc
3. Battery
Configuration &
Health
Battery Configuration, Voltage,
mAh, Technology,No. of Banks
Etc Along With Battery Health
Monitoring
4. Radio
Configuration
Wireless/Radio Configuration of
The Node/Device Such As
Transmit power, Receive
Sensitivity, Noise Floor Etc
Table: 3.2 Device Parameters
3.2 Trusted Node RF Ranging Based
Location/Distance Approximation
Figure: 3.2 Trusted Node RF Ranging Based
Location/Distance Approximation
Apart from theauthenticationpathdescribedabove,
another method of improving authentication score is
provided, for User/Device denied access by marginal
difference in authentication score, by verification of their
physical/geographic proximity to the trusted/central node
using a novel technique developed using Artificial
Intelligence & slowly increasing Transmit Distance by
varying the Radio Power using Transmit Power Estimation
ANN, until ping is replied by the User/Device requesting
access. After each ping, the pin generator increases the
power by a certain amount until a node connects. Once a
node does connect, the trained network calculates its
distance from an access controller and provides a key and
some leniency in access. The access controller will providea
false (fake) key if the neural networks calculated range is
greater than the coverage range. This is how we'll verify the
authenticity.

3.3 Artificial Neural Network Flow Chart
Figure: 3.3 Artificial Neural Network Flow Chart
Neural networks are self-learning adaptivesystems
that use interconnectednodes(sometimescalledneurons)in
a multi-tiered architecture likethehuman brain(alsoknown
as artificial neural networks). By being fed enough
information, a neural network can learn to identifypatterns,
organize information into categories, and make predictions
about the future. The input is broken down into levels of
abstraction by a neural network. It may be taught to detect
patterns in speech or images, for example, using many
samples, just like a human brain. A system's behavior is
determined by the way its components are linked together
and the weights attached to those links. To achieve the
required job, these weights are automatically modified
during training based on a predetermined learning rule. In
this flow chart we can see 1st our ANN code is start than call
to test values then input and targets test values then hidden
layer operation perform then process some functions than
some mathematical calculation will be perform then train
function, then plot error and performance plot after that
code will be stop.
4. RESULT
4.1 ANN Results
Figure: 4.1 Best Validation Performance Plot
The accompanying diagram depicts the Artificial
Neural Network's bestvalidationperformanceplot. The best,
validation, test, and railway lines are all displayed here.
Error typically reduces with additional training epochs but
may increase on the validation data set ifthe network begins
to overfit the training data. Epochs with low validation
errors get the greatest results with the default configuration
after a series of six consecutive increases in validationerror.

Figure: 4.2 Error Histogram Plot
The error histogram plot produced by our neural
network simulationisdisplayedabove.An errorhistogramis
a histogram of the deviations from the desired value, or
forecast, that occurred after training a feedforward neural
network. Because they indicate how far the projectedvalues
deviate from the desired ones, these error numbers can be
negative. The number of discrete categories (or "bins") in a
bar chart. The histogram is a form of bar chart that displays
numerical data by organizing it into discrete buckets. The
error bars in the preceding graph range from a high of
0.01913 at 7 instances to a low of 0.2843.
4.2 Estimated Power Results
Figure: 4.3 Ping Request- Distance & Estimated Power
The above result is for power estimation based on
distance, access controller starts sending ping requestatthe
0 second, and continuously sent pin with time interval of 10
seconds with increment of some power after each 10
seconds of time interval. At the 0 (zero) meter transmit
power estimated is 3 mw and ping is success after 100
seconds. At the distance of 101 meter and power estimated
on this distance is 20 mw ping is success. This distance is
outside the range of coverage area so there is risk in privacy
if connection is established so access controller sent fake
random key that is in the form of zero for provide securityto
network.
S. NO. Time Distance Power
1 0 Sec. 0 meter 3 mw
2 10 Sec. 10 meters 2 mw
11 100 Sec. 101 meters 20 mw
Table: 4.1 Ping Request- Distance & Estimated Power
Figure: 4.4 Fake Random Key Generated

In this above figure we can see ping received at the
distance of 101 meter and at this distance transmit power is
20 mw, this range of distance estimated is outside of the
allowed range, so access controller generates fake random
keys in the form of zeros and provide to node for
authentication and privacy of network.
4.3 Node Distance Based Access Controller Result
Figure: 4.5 Ping Request – Distance & Estimated Power
Above figureshowsdistance estimationwithpower,
access controller starts sending ping request with 2mw
power, at this power distance estimated by trained network
is 1 meter at 1second. Access controller sent ping request
continuously with increment of some power, after 19
seconds with 2 mw power estimated distance calculated by
trained network is 19 meters at this point ping is success
because this range in converge area so access controller
provide initial key for establish connection with some
leniency in access.
2 2 Sec. 2 meters 2.5 mw
5 8 Sec. 8 meters 2.5 mw
Table: 4.2 Ping Request – Distance & Estimated Power
Figure: 4.6 Initialized Key
In this above figure we can see estimated distance
and initialize key. Above figure shows that node is in
converge range, so node distance-based access controller is
ready for establish connection between node and access
controller, because ping is received at distance of 19 meter.
So, access controller proved initial key, generated initial key
we can also see in above figure.

Figure: 4.7 Generated Random Key
In this above figure we can see generated truly random key.
Figure: 4.8 Ping Request- Distance & Estimated Power
The above figure shows distance estimation with
power, access controller starts sending ping request with 0
mw power, at this power distance estimated by trained
network is 0 meter at when time is 0 second. Access
controller sent ping request continuously with increment of
some power with the time interval of 10 seconds, after 78
seconds with 15 mw power estimateddistancecalculatedby
trained network is 77 meters at this point ping is accept. But
this distance is not in converge range, so access controller
provides fake key for privacy of network.
Table: 4.3 Ping Request- Distance & Estimated Power
Figure: 4.9 Fake Random Key Generated

In this above figure we can see ping received at the
distance of 77 meter and their transmit power is 15mw,this
range of distance estimated is outside of the allowed range,
so access controller generates fake random keys in the form
of zeros and provide to node for authentication and privacy
of network.
5. CONCLUSIONS & FUTURE SCOPE
Conclusion
Blockchain technology is used in the suggested
system. Blockchain-based security mechanisms may also be
implemented in the real world. The same blockchain
architecture allows players to identify each device. The data
put into the system is unquestionable and may be used to
identify the device that supplied it. IoT applications will
benefit greatly from blockchain's ability to provide secure,
distributed authentication and authorization of IoT devices.
Smart contracts can confirm device message exchanges on
the blockchain, making it possible to secure device
communication.
IoT protocols can be more secure with the aid of
blockchain technology. In the Blockchain, messages sent
between devices may be seen as transactions, and smart
contracts can be used to verify them. IoT protocols can be
more secure with the aid of blockchain technology. The goal
proposed work is provided privacy in network based on
distance and power estimation, if any node tries to establish
connection in our network, if this is in our range then our
access controller provided true key for establish connection
with some leniency otherwise if this is in outside the range
then access controller providedfakerandomkeyforprivacy.
Several factors, including Path loss is the decrease in power
density of a radio signal over a long distance. Path loss is
more likely due to the gradual attenuation of radio signals
over their extended journey.
The so-called inverse square law states that the
radio wave power density increases as the square of the
distance decreases. The sensitivity of the receiver is the
second most essential factor in establishing a range. Fading
margin and environmental circumstances are two other
variables that must be accounted for in any computation.
There is also a potential height effect on the measuring
range, so be mindful of that while setting up your antenna.
The accompanying graph depicts nonlinearity that iscaused
by the nonlinear effects of several different variables in the
RF spectrum.
Future Scope
Biometric identifiers, fingerprints, voice samples,
retina or iris information, and other forms of biometric
identification all can be done for improvement in terms of
security. A key's dependability may also be estimated using
trust metrics built on top of the data records that make up
blockchain. There will be a lot of testing of Io Chain’s
performance and sturdiness through the development of
various apps on top of it. Also, the private Ethereum
blockchain network has to be updated so that the PoS-based
version of the ledger may be utilized, when it is publishedby
Ethereum developers. To expand this work into healthcare,
hospitality, pharmaceutical and education is possible in the
future.
REFERENCES
[1] Achraf Fayad, Badis Hammi, Rida Khatoun “A
Blockchain-based Lightweight Authentication Solution
for IoT” Cyber Security in Networking Conference2019.
[2] Aissam Outchkoucht, Hamza, Jean Philippe Leroy
“Dynamic Access Control Policy based on Blockchain
and Machine Learning for the Internet of Things”
International Journal ofAdvancedComputerScienceand
Applications 2017.
[3] Soumyashree S. Panda, Utkalika Satapathy,BhabenduK.
Mohanta “A Blockchain Based Decentralized
Authentication Framework for Resource Constrained
IOT devices” IEEE 2019.
[4] Khadija Fazal1, Adeel M.Syed “Blockchain
Authentication Mechanism for Securing Internet of
Things” PakJET 2020.
[5] Dr. Sakthi Kumaresh, Dr. K B Priya Iyer “Decentralised
Artificial Intelligence Enabled Blockchain Network
Model” Turkish Journal of Computer and Mathematics
Education 2021.
[6] Praveen Kumar Kollu, Monika Saxena, Khongdet
Phasinam “Blockchain Techniques for Secure Storageof
Data in Cloud Environment” Turkish Journal of
Computer and Mathematics Education 2021.
[7] Shanshan Zhao, Shancang Li “Blockchain Enabled
Industrial Internet of Things Technology” IEEE 2019.
[8] Muhammad Tahir, Muhammad Sardaraz , Shakoor
Muhammad “A Lightweight Authentication and
Authorization Framework for Blockchain-Enabled IoT
Network in Health-Informatics” Sustainability 2020.
[9] Dr Mohd Javaid, Prof. Abid Haleem, DrRaviPratapSingh
“Blockchain technology applications for Industry 4.0: A
literaturebased review” Journal Pre-proof 2021.
[10] Bhabendu K. Mohanta, Anisha Sahoo, Shibasis Patel
“DecAuth: Decentralized Authentication SchemeforIoT
Device Using Ethereum Blockchain” IEEE 2019.
[11] Jesse Yli-Huumo, Deokyoon Ko, Sujin Cho “WhereIs
Current Research on Blockchain Technology? —A
Systematic Review” PLOS ONE 2016.

Design of a Hybrid Authentication Technique for User and Device Authentication for an integrated IOT environment using Blockchain Authentication and Artificial Intelligence

More Related Content

Similar to Design of a Hybrid Authentication Technique for User and Device Authentication for an integrated IOT environment using Blockchain Authentication and Artificial Intelligence (20)

More from IRJET Journal (20)

Recently uploaded (20)

Design of a Hybrid Authentication Technique for User and Device Authentication for an integrated IOT environment using Blockchain Authentication and Artificial Intelligence