Edge-empowered accurate urban vehicle localization with cellular-aware trajectories

Hongzi Zhu, Fan Wu, Siyuan Cao, Shan Chang, Li Lu
2018 CCF Transactions on Networking  
Acquiring accurate vehicle location information in urban settings is very challenging due to the complexity of urban environments. In this paper, we propose a novel scheme, called UPS, to tackle urban vehicle localization problem. After extensive empirical study, we find that GSM power spectrogram collected over a distance has ideal temporal-spatial characteristics for fingerprinting. Encouraged by this observation, UPS tries to utilize the geographical trajectory and the associated GSM power
more » ... ectrogram information of a moving vehicle to identify its location with reference to a map. To this end, two appealing techniques, i.e., online vehicle localization and GSM map construction, are elegantly integrated. With the former, a vehicle can accurately fix its location under complex urban environments. With the latter, a reliable metropolitan-scale GSM power map can be cost-efficiently built at edges, leveraging the strong power of crowdsourcing. By design, UPS is light-weight, requiring only a minimum hardware deployment. We implement a prototype system to validate the feasibility of the UPS design. Furthermore, we conduct extensive trace-driven simulations and results show that UPS can work stably in various urban settings and achieve an accuracy of 5.3 m with a 90% precision, overwhelming the performance of GPS by five times.
doi:10.1007/s42045-018-0005-y fatcat:srljfrunmfdrtgprpngqqwg4cq