Design and Evaluation of an Indoor Positioning System based on Mobile Devices

Indoor Positioning Systems (IPS) using Bluetooth Low Energy (BLE) technology are very popular and use low-cost solutions. However, they usually suffer from poor proximity detection accuracy due to the reliance on Received Signal Strength Indicator (RSSI) that is prone to drastic fluctuations in the indoor environment due to interference and reflections. To mitigate these effects, a realistic Indoor Positioning System, based on typical BLE beacons and mobile devices that improves accuracy through a low-cost, one-time offline calibration process, is proposed and imple-mented. Indoor positioning techniques are enhanced by combining the filtering of both RSSI values and estimated positions while using complementary technologies such as QR and NFC to re-calibrate the system with exact starting positions. Comprehensive experimental studies on a real indoor environment with different system settings are conducted. The results demonstrate that the implemented approach achieves a meter accuracy in 87.2% of the estimated position values for the specified scenario.