Performance Characterization of Positioning in LTE Systems

This paper presents low-level models of cellular LTE signals along with an architecture for a software-defined radio (SDR) that is capable of (1) extracting the essential parameters for navigation from the received LTE signals and (2) acquiring and tracking real LTE signals transmitted from multiple base stations. Moreover, a method for obtaining time-of-arrival (TOA) estimates that is highly robust against interference and noise in the environment is presented. This method relies on estimating the channel impulse response from the received LTE signal. The proposed LTE SDR is evaluated on a ground vehicle in an urban environment with real LTE signals. Experimental results show a mean distance difference of 11.96 m between the proposed LTE SDR and the GPS solutions over a 1.42 Km with a standard deviation of 6.83 m and a maximum distance difference of 40.42 m.