Proposal of Adaptive Data Collection Method of On-Vehicle Sensors in the 5G Era

In the 5G era, the cellular network is designed to address various requirements from vertical industries, such as massive connectivity, high data rate and low-latency communication. Vehicle to everything (V2X) communication is considered a representative application in vertical domains. Vehicles and infrastructure can cooperate to develop a wide range of perception where sensor information from vehicles and roadside infrastructure is collected and processed at mobile edge servers. This perception information can be distributed for advanced driving support. For example, a receiver vehicle or driver extends local perception even in limited visibility. In this system, it is a key challenge to collect sensor information within the delay budget in order to provide real-time driving support. However, user experienced throughput may vary due to vehicle mobility. The quantity of sensor data that can be transmitted within the delay budget also varies. Therefore, in this paper, we propose a sensor data collection scheme with priority control and adaptive sensor set selection to maximize the quantity of sensor data under latency requirements. The study concludes that the proposed scheme is validated in a stationary environment based on 5G trials and laboratory tests.