An Intelligent Mechanism for Monitoring and Detecting Intrusions in IoT Devices

Abstract

As of recent years, the growth of data processed by devices has been exponential, resulting of the increasing number of Internet of Things devices connected to the Internet, which has come to play a very critical role in many domains, such as smart infrastructures, healthcare, supply chain or transportation. Despite its advantages, the amount of IoT devices has come to serve as a motivation for malicious entities to take advantage of such devices. To deal with potential cyberattacks in IoT devices, Machine Learning techniques can be applied to Intrusion Detection Systems along with Federated Learning to help manage privacy related concerns. Several intrusion detection methods have been proposed in the past, however, there’s a lack of research aimed at HIDS. Furthermore, the focus is mostly on applied ML methods and evaluation and not on real-world deployment of such systems. To tackle this, this work proposes a framework for a lightweight host based intrusion detection system based on system call trace analysis for benign and malicious activity detection. In summary, this work aims to present research about Host Intrusion Detection that could be applied for IoT devices, while leveraging Federated Learning for model updates.