Bluetooth Based Indoor Positioning Using Machine Learning Algorithms

摘要

Recently indoor positioning system (IPS) has been getting a lot of attention. Similarly, because of low cost and ease of use, Bluetooth low energy (BLE) is extensively used for IPS. Techniques such as trilateration, triangulation, and fingerprinting are widely studied in IPS. Fingerprinting is popular approach in RSSI based-IPS, but is also time-consuming method. Here, we proposed to use the state-of-the-art machine learning approach for fingerprinting. This paper proposes a BLE based machine learning location and tracking system for indoor positioning. The experimental results showed that the proposed method has an average estimation error of 50 cm.