Enhancing Security in Smart Buildings: Traffic Classification for Automated Access Control

The advent of the Internet of Things (IoT) has revolutionized the concept of Smart Buildings by integrating various smart sensing and control devices to improve efficiency and user experience. This thesis explores the realm of Smart Buildings, with a focus on the challenges of access control. In particular, the research investigates the applicability and effectiveness of machine learning techniques to classify device activities within the Smart Building environment. As Smart Buildings continue to evolve, the need to ensure robust security measures becomes paramount. Traditional, intrinsically static access control methods often struggle to adapt to dynamic environments. This study explores the feasibility of using machine learning algorithms to dynamically classify device activities, with the aim of improving access control mechanisms. Research leverages existing datasets to assess the robustness and accuracy of machine learning traffic classification by analyzing the traffic patterns generated by IoT devices. Machine learning models are then applied to classify these patterns into specific activities. By understanding and classifying different device activities, the system can dynamically adjust access permissions, contributing to a more adaptable and responsive security infrastructure. Preliminary results demonstrate that this type of approach has considerable potential, indeed, it has been possible to correctly classify about 98% of packets in network traffic from several IoT devices. The results of this research provide valuable insights into the field of IoT-enabled Smart Buildings, shedding light on the potential of machine learning to promote access control strategies. The implications of implementing such techniques go beyond security, impacting the overall functionality and sustainability of Smart Building environments.