Cecílio, José Manuel da SilvaFerreira, Pedro Miguel Frazão FernandesApolónia, Frederico José Cristovam2022-06-282022-06-2820222021http://hdl.handle.net/10451/53534Tese de mestrado, Engenharia Informática (Arquitetura, Sistemas e Redes de Computadores), 2022, Universidade de Lisboa, Faculdade de CiênciasEnergy consumption in the European Union has been growing above expectations, where buildings are the biggest consumers, ahead of the industry and transport sectors. Since the build ings are essential, the need arose to create a system that contributes to reducing energy costs. The Self Assessment Towards Optimization (SATO) proposes to interconnect several devices to collect data generated by buildings and send it to the SATO platform for self-assessment and optimization of energy resources inside the buildings. The location of people inside buildings can be done through cameras, sensors or signal strength, where the system’s intrusion varies. In this thesis, an inside location system for buildings was de signed and implemented. It uses Scanners Bluetooth Low Energy (BLE) to identify BLE devices without requiring users to install any application on their mobile phone or need to use some spe cific hardware. The data obtained from the probe of devices is sent to the SATO Platform, and then it is used to estimate the device location. The proposed solution is easily adjusted to any space since it does not require the fingerprinting for the new areas where the system will be imple mented. To achieve this goal, the data vectors obtained from the probing of devices are introduced into our data processing pipeline, which uses previously trained models. For creating the models, the space considered was divided into sub-spaces, where each sub-space corresponds to a geomet ric figure. These geometric figures are used to abstract new areas. Models for different geometric figures were created to validate the work. The Main Library of the Faculty of Sciences was used as a use case, where the proposed approach was used to estimate the device location. Results are promising, which shows that the prototype is functional and can be used to estimate the number of people in a given space.porOcupaçãoLocalização dentro de edifíciosBLEScannersTeses de mestrado - 2022master thesis202994848