Surface Correlation based Localization Technology using only LTE Signals in Urban Canyon

Abstract

In a dense urban area, availability of GNSS signal is very low, so it is very difficult to estimate the accurate location. As an alternative to this, we propose a surface correlation based fingerprinting method to improve the accuracy and reliability of the positioning performance using only LTE signals in urban canyon. The conventional fingerprinting methods utilize instantaneous RSS measurements from surrounding RF sources to compute a location of an object. This method shows very inaccurate performance in environments where the number of RF signal sources is small or noise and distortion of RF signals are severe. We solve this problem by using accumulated RSS pattern information during movement rather than instantaneous RSS information. The accumulated RSS pattern, named Surface, is not only robust to signal noise and distortion caused by the surrounding environment, but also improves the correlation accuracy with fingerprinting database. These advantages not only improve positioning accuracy and stability, but also maintain high accuracy even with a very small number of RF signal sources. In order to verify the proposed fingerprinting method, we have configured a RSS fingerprinting database of LTE signal in the old town of Barcelona. The target area is 57,694 square meters, and the availability of GNSS signals is very poor due to the densely distributed buildings. Through several tests, we verified the performance of the proposed method as a degree of 3 m accuracy. We compared the performance of the proposed method with the conventional fingerprinting methods and GNSS to verify the robustness, stability and accuracy are significantly improved. Through these results, it is confirmed that the proposed method can provide precise location information using only LTE signals in urban canyon environment.