Talk 1: Time series based real-life energy assessment of household appliances
Abstract: The European Union’s regulations for appliance energy labeling fall short of representing household appliances’ real-life energy efficiency and performance. A real-life energy performance assessment is essential to accurately evaluate energy utilization in daily usage conditions. By understanding better how real-life environments and usage patterns impact energy efficiency and appliance performance, users can make more informed choices and adjustments, and systems can promote better energy resource management in buildings. This thesis aims to: i) implement experimental facilities to collect time series data related to household appliances operation; ii) create a realistic and standardized definition of real-life appliance assessment; iii) define and implement a time series-based framework for online real-life assessment of appliances, and; iv) validate and evaluate the assessment framework. These are essential steps to overcome the limitations of current energy labeling regulations and contribute to improved integration of appliances in buildings’ energy management. In the scope of the European Union H2020 SATO project, the work will be carried out in close collaboration with energy sector experts, an appliance retail store, and a set of household owners/occupants providing access to appliance operational sensor data. The main objective of this thesis is to design and implement a robust and reliable time series-based frame- work that can accurately assess an appliance, from determining the type of appliance to identifying it and providing real-life energy efficiency and performance indicators. A suitable time series pipeline architecture is envisioned to effectively collect, analyze, and perform the necessary time-series classification and machine-learning tasks that will deliver the complete appliance assessment. The thesis results and conclusions will be evaluated and discussed with experts in the energy sector and delivered to a European Union agency involved in energy innovation and research.
Talk 2: LWSEE: Lightweight Secured Software-Based Execution Environment
Abstract: As the Internet of Things (IoT) continues to expand, data security has become increasingly important for ensuring privacy and safety, especially given the sensitive and, sometimes, critical nature of the data handled by IoT devices. There exist hardware-based trusted execution environments used to protect data, but they are not compatible with low-cost devices that lack hardware-assisted security features.
The research in this paper presents software-based protection and encryption mechanisms explicitly designed for embedded devices. The proposed architecture is designed to work with low-cost, low-end devices without requiring the usual changes on the underlying hardware. It protects against hardware attacks and supports runtime updates, enabling devices to write data in protected memory.
The proposed solution is an alternative data security approach for low-cost IoT devices without compromising performance or functionality. Our work underscores the importance of developing secure and cost-effective solutions for protecting data in the context of IoT.