Vehicle localization is essential for intelligent transportation. However, achieving low-latency vehicle localization without sacrificing precision is challenging. In this paper, we propose a road-aware localization mechanism in heterogeneous networks (HetNet), where distinct features of HetNet signals are extracted for two-spatial-scale position mapping, enabling low-latency positioning with high precision. Specifically, we propose a sequence segmentation method to extract the low-dimensional positioning space on two spatial scales. To represent roads and sub-segments according to HetNet signals, we propose a salient feature extraction method to eliminate redundant features and retain distinct features, thereby reducing feature-matching complexity and improving representation accuracy. Based on the extracted salient features, a two-spatial-scale localization algorithm is designed through salient feature matching, which can achieve low-latency road-aware localization. Furthermore, high-precision positioning is achieved by coordinate mapping based on curve fitting. Simulation results show that our mechanism can provide a low-latency and high-precision positioning service compared to the benchmark schemes.
@article{arxiv.2312.15443,
title = {Road-Aware Localization With Salient Feature Matching in Heterogeneous Networks},
author = {Lele Cong and Deshi Li and Kaitao Meng and Shuya Zhu},
journal= {arXiv preprint arXiv:2312.15443},
year = {2023}
}