A New Location Algorithm Based on Machine Learning with Fingerprinting in C-V2X

The location algorithm in C-V2X is one of the important technical approaches for the development of autonomous driving services. Currently, it is based on many positioning solutions such as base stations and satellites. In autonomous driving positioning scenario, challenges such as positioning accuracy, processing delay, and deployment cost are often encountered. This paper proposes a new fingerprint location algorithm based on RSU in C-V2X, which solves insufficient signature parameter and road grid problems encountered when traditional fingerprint positioning technology is directly applied. Meanwhile, a new polar coordinate fingerprint library used in location algorithm is constructed to be suitable for C-V2X scenarios. In addition, a new method of sub-fingerprint library is used to improve the positioning accuracy and handover effectiveness of fingerprint positioning which provides an effective positioning method for C-V2X scenarios.