Accurate Localization Using LTE Signaling Data

In this paper, we propose a novel localization algorithm using LTE signaling data. Specifically, we use TA (Timing Advance) and RSRP (Reference Signal Receiving Power) data that are required in LTE standard and already available in current LTE systems. The combination of (TA, RSRP) is used as a signature, and one can expect that different locations will have distinctive signatures. Our real world data first showed that both TA and RSRP at a particular location have a small variance, and therefor good for localization purpose. We then consider the spatial-temporal dependencies of these data to improve the localization accuracy. We design and implement a novel machine learning based algorithm to infer a user's latitude-longitude given a (TA, RSRP) measurement. We have evaluated our system using traces collected in our city covering more than 800 km. The results show that our cellular signaling based localization algorithm achieves a median localization accuracy around 25 m.