INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 9, ISSUE 03, MARCH 2020 ISSN 2277-8616

Classification For Iot Threats Based On The

Abstract: The Internet of Things technology make us depending on a machine to control a massive part of our life by collecting more personal
information. The data collected and stored with these devices such as your name, age, health data, location and more can aid cyberattack activity. The
first step to face these threats is to classify it and determine the risk according to different classes of users. This paper introduces classification to IoT
threats from a user perspective. The classification is done by collecting the opinion of 80 users divided into three classes. Our proposed classification is
done by AHP algorithm to calculate the weigh and determine the risk of each threat according to users classes. The result will contribute to create more
secure and reliable IoT services and systems.

Index Terms : analytic hierarchy process, Internet of Things, Quality of service, security risk, Threat.
——————————  ——————————

1 INTRODUCTION Uday Kumar, and Sugata Sanyal have surveyed all security
A new chapter in the future of humanity started when the flaws, along with an analysis of the privacy issues that an end-
Internet of Things (IoT) technology invaded our life. We live in user may face. They concluded that more security measures
an era that guided by computers, which collect data from our should have existed before implementing IoT systems in our
surrounded environment and then serve us according to it. daily life [4]. In [5] Jyoti Deogirikar and Amarsinh Vidhate
Nowadays, most devices such as televisions, thermostats, discuss the problem of building a secure IoT system.
sensors in your car, and medical devices are finding their way Implementing different security measures could consume
online without our help. Many refer to this as the "Internet of more power, which is not practical, because of that, there is a
Things". The concept behind IoT idea involves a variety of need for a security mechanism that handles maximum security
objects that can be connected using either wireless or wired problems and be lightweight for robust IoT. In [6] Yang Lu, and
networks. These objects have a unique addressing scheme Li Da Xu Discuss what people need from IoT technology and
that allows them to interact and cooperate to create novel IoT cybersecurity. The complex structure of IoT led to
applications and services [1]. According to IoT security report comprise data model, interfaces and protocol. In fact, we do
in 2018 from National Institute of Standers and Technology not have a standardised framework that can integrate data
(NIST) 25% firms reported security issues in IoT which lead to models, ontology, and data formats with IoT protocols,
damages cost of at least 34 million dollars in the last couple of applications, and services. As a result, QoS (Quality of
years. This number will increase with more security threats service) research is needed to support the development of IoT
arising every day, which led to different classification and Data privacy and integrity. In [7] Mirza Abdur, Sajid Habib,
complex security measures [2].Ideally, one would like to have Muhammad Ali and Saleem Ullah have categorised twelve
a single classification for these threats. However, different threats to low-level attacks, medium-level attacks, high-level
systems have different use and users. Also, most classification attacks, and extremely high-level attacks according to their
is focused on threats nature but not the system users. It will be effect and behaviour. The suggested solution will be difficult to
unpractical to add more security layers to IoT massive system apply due to the different behaviour of threats and the
without gaining the user trust to use the system. The core of complexity of IoT. In [8] Se-Ra Oh and Young-Gab Kim
this study is to classifies IoT threats by determining the risk analysed three essential characteristics heterogeneity,
according to the opinions of different users. To do that First, resource constraint and dynamic environment to find out basic
we describe the IoT security threat and classify them. Then, IoT security requirements. Based on six elements IoT network,
we collect the data needed for calculation using two cloud, user, attacker, service and security requirements
questionnaires. After that, the AHP algorithm is applied to find evaluation is performed to be used as a guide to design
the risk for every threat. Finally, a conclusion is presented to secure IoT systems. In [9], Peter Aufner discuss the gap
summarise the main outcomes of this research. between threat modelling and frameworks in IoT security. By
quantitative three threats modelling framework STRIDE,
2 RELATED WORK LINDDUN and CORAS three gaps in security research
The Internet of Things is one of the fastest-growing industry founded. The first gap is between threats modelling
and technology in the world of wireless Network frameworks and IoT because frameworks developed during a
Telecommunication. The main objective of this technology is to time when software could examine regard of hardware. The
provide connectivity between heterogeneous network devices second gap is between threats modelling frameworks and
[3]. Tuhin Borgohain, common security research. Here, most research focuses on
frameworks without having a deep explanation of threats
____________________________________ modelling and possible attacks, which led to weakness and
confusing standards for IoT security. The third gap is between
• Islam Abdalla Mohamed, Department of Computer Science, Jouf security research and IoT itself. Most research focuses on IoT
University, Saudi Arabia, [email protected].
software and network regards of the attacks that rise with new
• Anis Ben Aissa, Department of Computer Science, Jouf University,
Saudi Arabia, [email protected]. devices deployed in IoT systems. In [10] Keon Chul Park and
• Loay F. Hussein, Department of Computer Science, Jouf University, Dong-Hee Shin proposed a new framework that integrates
Saudi Arabia [email protected]. fuzzy DEMATEL and fuzzy ANP to reflect dependence and
4860
IJSTR©2020
www.ijstr.org