EAI Endorsed Transactions
on Internet of Things Research Article
1
Detections of IoT Attacks via Machine Learning-Based
Approaches with Cooja
Ali Hamid Farea
1,*
, and Kerem Küçük
2
1
Department of Computer Engineering at Kocaeli University, Kocaeli, Turkey
2
Department of Software Engineering at Kocaeli University, Kocaeli, Turkey
Abstract
Once hardware becomes "intelligent", it is vulnerable to threats. Therefore, IoT ecosystems are susceptible to a variety of
attacks and are considered challenging due to heterogeneity and dynamic ecosystem. In this study, we proposed a method
for detecting IoT attacks that are based on ML-based approaches that release the final decision to detect IoT attacks.
However, we have implemented three attacks as a sample in the IoT via Contiki OS to generate a real dataset of IoT-based
features containing a mix of data from malicious nodes and normal nodes in the IoT network to be utilized in the ML-based
models. As a result, the multiclass random decision forest ML-based model achieved 98.9% overall accuracy in detecting
IoT attacks for the real novel dataset compared to the decision tree jungle, decision forest tree regression, and boosted
decision tree regression, which achieved 87.7%, 93.2%, and 87.1%, respectively. Thus, the decision tree-based approach
efficiently manipulates and analyzes the KoÜ-6LoWPAN-IoT dataset, generated via the Cooja simulator, to detect
inconsistent behavior and classify malicious activities.
Keywords: IoT security, Attacks, Machine Learning-based approaches, Decision tree-based models, Cooja simulator.
Received on 01 March 2022, accepted on 02 April 2022, published on 07 April 2022
Copyright © 2022 Ali Hamid Farea et al., licensed to EAI. This is an open access article distributed under the terms of the Creative
Commons Attribution license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work
is properly cited.
doi: 10.4108/eetiot.v7i28.324
1. Introduction
An Internet of Things (IoT) is a network of physical objects
containing sensors, actuators, microcontrollers, and smart
appliances that gather and transfer information and interact
with their surroundings [1], [2], allowing these devices to
generate and exchange data with minimal human
intervention. It is one of the most promising technologies and
the world is already beginning to utilize various IoT
technologies. It communicates with each other via various
protocols [3] as well as interacts with a wide range of
applications, including smart cities, building automation,
safety, surveillance systems, logistics, healthcare, economy,
calamity and agriculture [4], [5], [3]. Therefore, it offers a
large number of attractive qualities that have made us rely on
it in our daily applications with best-effort and real-time [6],
[7].
*
Corresponding author. Email: 195112025@kocaeli.edu.tr
The IoT cloud provides capabilities for collecting,
processing, managing, and storing massive amounts of data
in real-time [8], [9]. This data may be easily accessed
remotely via industries, governments, monitoring tools, and
related services, allowing them to make decisions as needed
[10], [11]. It is essentially a powerful, high-performance
network of servers designed to do high-speed data processing
for billions of connected devices [12].
IoT technologies have certain properties in common that
are described as heterogeneity, auto-configuring, dynamic
ecosystem, smart, large scale, and connectivity [4], [13], [14],
[15]. For example, the IoT ecosystem includes extremely
different technologies and protocols, adaptive protocols, a
variety of factors that may be influenced in order to adapt to
environmental changes, etc. These components (large scale)
work together in a cooperative and smart way to share their
collected data and services [16]. In many cases, the connected
devices are required to offer secure and reliable services to an
applicant [17].
EAI Endorsed Transactions on
Internet of Things
04 2022 - 04 2022 | Volume 7 | Issue 28 | e1