Hybrid Machine Learning for IoT-Enabled Smart Buildings

This paper presents an intrusion detection system (IDS) leveraging a hybrid machine learning approach aimed at enhancing the security of IoT devices at the edge, specifically for those utilizing the TCP/IP protocol. Recognizing the critical security challenges posed by the rapid expansion of IoT networks, this work evaluates the proposed IDS model with a primary focus on optimizing training time without sacrificing detection accuracy. The paper begins with a comprehensive review of existing hybrid machine learning models for IDS, highlighting both their strengths and limitations. It then provides an overview of the technologies and methodologies implemented in this work, including the utilization of “Botnet IoT Traffic Dataset For Smart Buildings”, a newly released public dataset tailored for IoT threat detection. The hybrid IDS model is explained in detail, followed by a discussion of experimental results that assess the model’s performance in real-world conditions. Furthermore, the proposed IDS is evaluated for its effectiveness in enhancing IoT security within smart building environments, demonstrating how it can address unique challenges such as resource constraints and real-time threat detection at the edge. This work aims to contribute to the development of efficient, reliable, and scalable IDS solutions to protect IoT ecosystems from emerging security threats.