Autonomous Vehicle Communication in V2X Network with LoRa Protocol


Yau Ka Cheung

Document Type


Degree Name

Master of Science (MS)


Computer Science and Info Sys

Date of Award

Spring 2020


In the Vehicle-to-Vehicle or Vehicle-to-Infrastructure (V2X), the communication stability is one of important concerns to improve safety of automatic vehicle. The weakness of Wi-Fi is short range and Long Term Evolution (LTE) is possible to lose connection or uncover in certain area. In this research, we proposed to take advantage of Long-Range (LoRa) protocol and Long-Range Wide Area Network (LoRaWAN) and focusing on gateway selection in order to improve the overall communication performance in the V2X system and data transmission [1]. The performance of communication can be enhanced by wide cover signals of LoRa and improved by our developed mechanism, YORA, for gateway selection and elimination in the LoRaWAN server. By obtaining the segmentation images from the map through trained deep learning model and then weighting the segmentation, connection possibility between vehicles and gateways can be calculated. In the research, we developed a 93% accuracy model to generate the map segments. Meanwhile, transmission latency between vehicles and infrastructures also can be reduced by minimizing the messages' payload content. As a result, we can utilize the resources better by using minimum number of LoRa gateways and maintaining highly possible stable connection.


Meikang Qiu

Subject Categories

Computer Sciences | Physical Sciences and Mathematics