Minimizing the Age of Information in Intelligent Transportation Systems

Abstract

In many applications offered by intelligent transportation systems (ITS), maintaining the freshness of real-time information is a key requirement for the successful delivery of their services. The age of information (AoI) is a new metric recently proposed to capture the data freshness. This paper considers intelligent vehicles in a Vehicle to Infrastructure (V2I) network where each vehicle has a stream of data sampled by on-board sensors to communicate with a road side unit (RSU). Many vehicles demand access and vehicles stay in range for a short period of time. The objective of the RSU, then, is to schedule transmissions of these vehicles with the objective of maintaining data freshness upon receiving these packets. We first formulate the scheduling problem as a mixed integer linear program (MILP) to minimize the weighted AoI, accounting for the dynamic nature of the environment (vehicles' arrivals and their speeds, channel reliability, etc.). We also propose a scalable greedy algorithm that solves the problem in a polynomial time, and we prove that it obtains the optimal solution. We generate synthetic data using SUMO and evaluate numerically the performance of our algorithms.