Enhancing Intrusion Detection in IoT Systems through Simulated Attack Scenarios
Marwa Neily, Farah Jemili, Ouajdi Korbaa · Journal of Advances in Information Technology · 2025
The Internet of Things (IoT) landscape is fraught with vulnerabilities, making it a prime target for various types of attacks.While existing literature has extensively explored IoT attacks through studies and simulations, this paper introduces a fresh perspective by proposing a new methodology for testing attacks in IoT environments.Focusing on six prominent attack vectors, we conduct comprehensive tests using both the Cooja and OMNET++ simulators.Our research delves into the underlying factors driving these attacks, analyzing data based on the attackers' chosen target addresses.Through our novel approach, we aim to deepen the understanding of IoT vulnerabilities and provide insights into the behavior of attackers, ultimately paving the way for more effective defense mechanisms in IoT ecosystems.Our experiments demonstrate high detection accuracy, with Random Forest achieving up to 99.86% accuracy in Cooja and 98.66% in OMNET++.Despite these results, the study acknowledges potential limitations in realworld generalization due to the simulated nature of the dataset.