Search

Indoor positioning has become popular in this decade and is used to locate users or objects in indoor environments. This is because global positioning system (GPS) is not efficient for indoor use due to the multipath fading effect. This research is about development bluetooth low energy (BLE) indoor positioning system with the aid of long range (LoRa) network and guideline on selection of the BLE beacons. Next, positioning systems are developed consisting of BLE beacons, a transceiver of hybrid BLE-LoRa module, a LoRa receiver and Raspberry Pi as real-time monitoring. The received signal strength indicator (RSSI) and BLE Mac address from BLE beacons received via LoRa network are analyzed using the positioning algorithm designed in MATLAB. The positioning algorithm incorporates distance estimation, filter implementation and trilateration technique. The estimated location is analyzed with the root mean square error (RMSE) and cumulative distribution function (CDF). According to the results, implementing the filter reduces the positioning accuracy error, achieving 90% accuracy of positioning error less than 1.20 meters for the whole testbed. Finally, the algorithm is embedded into Raspberry Pi to view the location via desktop.

Indoor positioning systems has become popular in this era where it is a network of devices used to locate people or object especially in indoor environment instead of satellite-based positioning. The satellite-based positioning global positioning system (GPS) signal is affected, and loss incurred by the wall of the building causes the GPS lack of precision which leads to large positioning error. As a solution to the indoor area coverage problem, an indoor positioning based on bluetooth low energy (BLE) and long range (LoRa) system utilising the receive signal strength indicator (RSSI) is proposed, designed, and tested. In this project, the prototype of indoor positioning system is built using node MCU ESP 32, LoRa nodes, and BLE beacons. The node MCU ESP 32 will collect RSSI data from each BLE beacons that deployed at decided position around the area. Then, linear regression algorithm will be used in distance estimation. Next, particle filter is implemented to overcome the multipath fading effect and the trilateration technique is applied to estimate the user’s location. The estimated location is compared to the actual position to analyze the root mean square error (RMSE) and cumulative distribution function (CDF). Based on the experiment result, implementing the particle filter reduces the error of location accuracy. The particle filter achieves accuracy with 90% of the time the location error is lower than 2.6 meters.