Comparação de métodos de localização de dispositivos em ambiente fechado conectados em uma rede bluetooth mesh

Abstract

The indoor localization of devices is a highly relevant topic in the context of the Internet of Things (IoT), especially given the growing market and the demand for precise and scalable solutions. This work compares various positioning methods using Bluetooth Mesh networks and examines the influence of parameters such as device layout, applied filters, and position determination algorithms. Multilateration and FingerPrinting techniques associated with neural networks were evaluated, as well as several filters for processing RSSI (Received Signal Strength Indicator). The experiments indicated that the combination of neural networks with an adaptive filter resulted in lower mean errors in determining device positions. The results contribute to the improvement of indoor localization systems by guiding the selection of more efficient configurations for IoT applications.