Exploring The Security Issues in Home-Based IoT Devices Through Denial of Service Attacks

Abstract

Internet of Things (IoT) has infiltrated the digital realm, and critical efforts are being made to create robust security for these technologies. However, with increasingly sophisticated attacks, it is essential to understand IoT device security in depth. To understand the device vulnerabilities from the network level, we orchestrated Denial of Service (DoS) attacks for four different IoT devices through network flooding. We conducted our experiment in the lab environment using other IoT devices that include the Amazon Echo, a smart light-bulb, a smart camera, and a smart garage door opener. To conduct the DoS attack, we used Raspberry Pi as the main target to access other network devices with different protocols. We generated the DoS attack using Kali Linux installed in a virtual environment. This experiment demonstrated that hackers might exploit sensor vulnerabilities to gain unauthorized network access and use user data through various IoT devices. We proposed an effective Intrusion Detection technique using a combination of machine learning classifiers and deep learning. Some of the machine learning models include the logistic regression, decision tree, random forest and support vector machine to detect and mitigate the attack. The outcomes show the algorithm which presents the highest degree of attack detection accuracy. Our findings also show that DoS attacks continue to be a significant concern even with improved technologies and security protocols. Finally, we provide design implications to address such critical security flaws.