A Protocol-Channel-Based Indoor Positioning Performance Study for Bluetooth Low Energy

• Main Authors: Gabriel De Blasio, Alexis Quesada-Arencibia, Carmelo R. Garcia, Jose Carlos Rodriguez-Rodriguez, Roberto Moreno-Diaz
• Format: Article
• Language: English
• Published: IEEE 2018-01-01
• Series: IEEE Access
• Subjects: Bluetooth low energy; indoor positioning; fingerprinting; distance and similarity measure; protocol; channel
• Online Access: https://ieeexplore.ieee.org/document/8360100/

Bluetooth low energy (BLE) technology coupled with fingerprinting provides a simple way to position users with high accuracy in indoor environments. In this paper, we study the effect of BLE protocols and channels on indoor positioning using different distance and similarity measures in a controlled environment. With the aim of reproducing a real positioning system situation, we also study the effect of the user's orientation in the positioning phase and, consequently, provide accuracy and precision results for each orientation. In a 168-m² testbed, 12 beacons configured to broadcast with the Eddystone and iBeacon protocols were deployed and 40 distance/similarity measures were considered. According to our results, in a specific orientation there is a group of distance metrics coupled with a protocol-channel combination that produces similar accuracy results. Therefore, choosing the right distance metric in that specific orientation is not as critical as choosing the right protocol and, especially, the right channel. There is a trend whereby the protocol-channel combination that provides the best accuracy is almost unique for each orientation. Depending on the orientation, the accuracies obtained for the above mentioned group of distances are within the range of 1.1-1.5 m and the precisions are 90% within the range of 1.5-2.5 m.