In this paper, we evaluate the ability of connected roadside infrastructure to provide traffic predictions on highways based on the motion of connected vehicles. In particular, we establish metrics to quantify the amount of traffic prediction that is available from roadside units via vehicle-to-infrastructure (V2I) communication. We utilize analytical and numerical tools to evaluate these metrics as a function of (i) the location of the roadside units along the road, (ii) the communication range of the roadside units, and (iii) the penetration rate of connected vehicles on the road. We show that considerable amount of traffic predictions can be achieved even with sparsely distributed roadside units as distant as two thousand meters and with connected vehicle penetration rate as low as 2%. Based on the proposed metrics, we develop strategies for deploying roadside units along highways so that traffic prediction efficiency is maximized. Ultimately, the results of this paper may serve as a guideline for evaluation and deployment of connected roadside infrastructure.

在本文中，我们评估了联网路边基础设施根据联网车辆的运动提供高速公路交通预测的能力。特别是，我们建立了指标来量化路边单元通过车辆到基础设施 (V2I) 通信可获得的交通预测量。我们利用分析和数值工具来评估这些指标作为 (i) 沿路路边单元的位置，(ii) 路边单元的通信范围，以及 (iii) 联网车辆在道路上的渗透率的函数。路。我们表明，即使在距离达 2 千米的稀疏分布的路边单元和低至 2% 的联网车辆渗透率的情况下，也可以实现大量的交通预测。根据建议的指标，我们制定了沿高速公路部署路边单元的策略，以便最大限度地提高交通预测效率。最终，本文的结果可作为评估和部署互联路边基础设施的指南。