Priority control in communication networks for accuracy-freshness tradeoff in real-time road-traffic information delivery

Delivering real-time road-traffic information to the driver is a straightforward solution to the problem of road-traffic congestion. The information is more effective as it is fresh and more accurate. However, real-time road-traffic information delivery has a fundamental problem: an accuracy-freshness tradeoff. Unfortunately, real-time road-traffic information delivery has difficulty satisfying both requirements. To guarantee the freshness, the information needs to be delivered on the basis of the data received by a cloud or edge server before a predetermined deadline. However, only a limited amount of data is received due to bandwidth limitation and processing overhead in communication networks, which results in the poor accuracy of the delivered information. The only way to improve the accuracy is to make the deadline less strict, which results in deteriorating the freshness of information. The proposed system solves this tradeoff. The key idea is that data more ‘‘important’’ for the accuracy of information are more prioritized when the data are transferred in communication networks. In the proposed system, ‘‘importance’’ is determined by how helpful the data are when the system needs to estimate missing spatial information from a limited amount of received data by using the machine learning technique. In this paper, simulation results verify that the proposed system ensures the accuracy of road-traffic information while satisfying the freshness requirement.