Sistema de posicionamento em ambientes internos baseado em bluetooth low energy com uso de filtragem estocástica e fusão de sensores

• Ano de defesa: 2025
• Autor(a) principal: Fabris, Andrey
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Pós-Graduação em Sistemas de Energia; UTFPR
• Palavras-chave em Português: Sistemas de posicionamento interno (localização sem fio); Kalman, Filtragem de; Tecnologia bluetooth; Algorítmos - Desenvolvimento; Desempenho - Avaliação; Python (Linguagem de programação de computador); Indoor positioning systems (Wireless localization); Kalman filtering; Bluetooth technology; Algorithms - Development; Performance - Evaluation; Python (Computer program language); CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/36835

This work proposes an indoor positioning system (IPS) that integrates multiple radio frequency techniques through a modified Track Fusion with Fused Prediction (TFP) algorithm. The employed signal metrics include Received Signal Strength Indicator (RSSI), Angle of Arrival (AoA), and a combination of both within the Bluetooth Low Energy (BLE) 5.1 framework. While AoA stands out for its precision, low energy consumption, and cost-effectiveness, RSSI is characterized by its simplicity and widespread availability. By applying these measures in algorithms such as multilateration (MLT), trigonometry, and triangulation, the coordinates of a target can be obtained. To validate the system, a dataset containing real RSSI and AoA measurements from a transmitter and four BLE 5.1 receivers in a 14 × 8 meter environment was utilized. The Kalman Filter (KF) was employed to improve the accuracy of multilateration, AoA combined with RSSI, and AoA-only algorithms. Moreover, one more step was considered in the IPS, where the aforementioned KF-filtered outputs are then fused through the modified TFP. Results demonstrate that the proposed scheme, which is referred to as Angle–RSSI Fusion Localization (ARFL), significantly improves localization accuracy compared to other techniques. In particular, it reduces up to 81,61% in the average position error when compared to multilateration with KF. This advanced IPS offers a cost-effective and precise solution suitable for various applications in industries such as healthcare, commerce, and logistics.