MRT trains mobility tracking using LoRa network

• Main Authors: Pasia, Samuel Alexander A., Rivera, Vince Matthew A., Sy, Jereme Adriane D.G., Tan, Bianca Clarisse Y.
• Format: text
• Language: English
• Published: Animo Repository 2022
• Subjects: Street-railroads—Philippines; Wireless communication systems; Internet of thing; Electrical and Computer Engineering; Electrical and Electronics; Systems and Communications
• Online Access: https://animorepository.dlsu.edu.ph/etdb_ece/5; https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=1004&context=etdb_ece
• Institution: De La Salle University

This study aims to develop a smart monitoring system for the Metro Rail Transit (MRT) system in the Philippines using the LoRa network. To achieve this, the researchers developed a sensor node consisting of the Arduino MKR1300, NEO6M GPS module, and the A3144 hall effect sensor. This sensor node sends packets that contain the position and speed of the train to the RAK7246 gateway. To test the system, they utilized two automotive vehicles. The researchers measured the RSSI values at increasing distances to determine the capabilities of LoRa when implemented in the MRT system. From here, they found out that the maximum distance with an RSSI value between -30 dBm to -120dBm is approximately 700m. Additionally, they also measured the RSSI values for different train simulations such as when two trains are parked in a station, two trains are going past each other and one train is leaving a station while the other is parked. The results show that adjacent nodes do not cause interference with each other as there is no effect on the RSSI values for both. Furthermore, the researchers developed a Passenger Information Display System (PIDS) which shows information about the trains such as its speed, position relative to the stations, and the estimated time of arrival to its destination. The PIDS incorporates different algorithms for different train scenarios such as switching lanes or going underground and losing GPS signal, etc. They were also able to measure the delay between the actual arrival of the train and the indicated arrival on the app; the results showed that the average delay was only 3.9 seconds indicating that the system is responsive.