ANALYSIS OF BLUETOOTH LOW ENERGY-BASED INDOOR LOCALIZATION SYSTEM USING MACHINE LEARNING ALGORITHMS

Based on location, positioning systems are generally divided into two groups of indoor and outdoor environments. In the indoor type, the location of the device or the user is obtained within an indoor setting, in which technologies such as Bluetooth are used. The current availability of those indoor positioning systems (IPSs) that employ the Bluetooth low-energy (BLE) approach rendered them to be more popular among users worldwide. Nevertheless, such technologies are still facing a number of issues, particularly those associated with the instability of the received signal strength indicator (RSSI). The present work targets the improvement of distance and the selection of the optimal algorithm for the classification of the positions based on the BLE technology. The main contribution of this work is finding the most accurate location finding algorithm to classify the position using Bluetooth with low energy technology. To achieve this, we employed a machine learning approach that involves the design of algorithms which facilitate the learning process by the computer. Following the use of a specific four-algorithms training dataset, we achieved the best result at 71% true classification. In future work, we will use the present outcomes in the field of mobile applications.