A Study of Filtering Method for Accurate Indoor Positioning System Using Bluetooth Low Energy Beacons

Fingerprinting technique is an essential element in the indoor positioning system (IPS). Common methods utilize Wi-Fi signals. However, most of the Wi-Fi, because it is pre-installed for other purposes, is difficult to fully used for fingerprinting. Bluetooth beacon fingerprinting (BBF) can be used to overcome the drawbacks of the Wi-Fi fingerprinting. With BBF cost reduction is possible, and it also has advantages that can flexibly correspond to the environment variation. This study determines whether it is possible to achieve IPS with implementation of BLE beacon-based fingerprinting. This study presents an optimized fingerprinting method for building a fingerprinting data. In order to increase the accuracy, we applied filtering and probabilistic smoothing technique of fluctuation correction to necessary part of the signal. Result of the study proposes that BLE beacon fingerprinting can be a replacement for Wi-Fi based fingerprinting.