Improved Accuracy for Indoor Positioning with Bluetooth 5.1: From Theory to Measurements

Sammanfattning: In this thesis we are evaluating Bluetooth Low Energy (BLE) v5.1 in combination with Direction Of Arrival (DOA) algorithms and receiving antenna arrays for indoor positioning systems. The thesis is divided in two parts. The first part is the core of the thesis and is composed by the evaluation of DOA algorithms in a direction finding system. A campaign of measurements is carried out in an active indoor human environment. We aim for a suitable antenna configuration and DOA algorithm that will result in an angular error estimation below 10. In the process, we evaluate the implications of the frequency in AOA estimation and we propose an algorithm that is based on frequency diversity in order to make angular estimation errors more robust. The objective is successfully complied using a 4x4 URA receiver together with PDDA DOA estimation algorithm reaching a Mean Absolute Error (MAE) of 4.15. The second part of the thesis consists of demonstrating that a sub-meter error accuracy is possible using Bluetooth v5.1. To do so, we use the findings from the first part of the thesis and we present a simple geometric approach to estimate the position of the transmitter. The positioning engine is based on AOA and RSSI parameters. The objective is again successfully completed, obtaining an average distance error of 0.88 m.