Estimação do Ângulo de chegada utilizando Bluetooth 5.1 e redes neurais profundas

Le résumé fourni par la source

Indoor positioning systems have been driven by the increased use of mobile devices and the need for precise location or orientation in various sectors. Bluetooth Low Energy (BLE) 5.1 stands out for spatial orientation tasks due to its low energy consumption and the availability of resources for angle of arrival (AoA) determination through IQ quadrature samples. The processing of these samples is crucial for AoA estimation. However, the literature does not yet provide machine learning models, specifically deep learning models, to estimate the angle of arrival based on these samples. In this context, this work proposes the use of deep learning techniques for AoA estimation through a regression model applied to Bluetooth-based indoor orientation systems. We present the entire data collection scenario and the necessary procedures for sample and proposed model validation. Using the Mean Absolute Error (MAE) performance metric, we observed an error of 1.38° in estimating the angle of 135°, and overall, the proposed model achieved an MAE of 1.87°.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.