Mobile Device Localization in 5G Wireless Networks

Abstract

As wireless networks are evolving into 5G, tremendous amount of data will be shared on the newly developed open source platforms. These data can be used in developing new services. Among which, location information of mobile devices are extremely useful. For example, the location information can be used to assist wireless operators to trouble shoot the network performance. It can also be used to provide some location assisted service. However, some of these devices may be designed for limited budget that do not have the capability of GPS. Furthermore, operators may not have access to the GPS information on the mobile devices. In this paper, we propose a novel machine learning based approach to estimate the location of the mobile devices based on the measurement data that mobiles reported during every call and session. Our proposed algorithm utilizes the advanced features of 5G wireless network, such as the beam information. Simulation shows that the proposed solution can achieve 4m accuracy for LoS enviorment and 12m accuracy for mixed LoS and NLoS environment. And the proposed algorithm can also work even with only the information from one base station.