Contents I. Introduction
The integration of sensor capabilities into wireless infrastructure is considered a major part of the evolution of the sixth-generation (6G) communications standard [1]. An extensive survey of localization sensor systems for autonomous driving is provided in [2]. Here, a differentiation is made between three categories of localization systems - namely, conventional localization, machine learning-based localization, and vehicle-to-everything (V2X) localization. The latter can be divided into vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) localization. A possible application of V2I localization is depicted in Fig. 1, in which a traffic light is equipped with a stationary wireless sensor that interacts with sensors mounted on the vehicles. The accurate localization of the road users in critical traffic situations, such as crossroads, is vital for enabling fully autonomous driving. Various sensor systems can be integrated into the infrastructure to enable V2I localization. In [3], several radio frequency identification (RFID) tags are mounted beside the road for vehicle localization, achieving a 2D positioning accuracy of up to 27 cm. The approach in [4] is also tempting because it uses an already established LTE-V communication channel but achieves a localization accuracy of only about 10 m, which is not sufficient in critical situations.
