Context-Aware Telco Outdoor Localization

Recent years have witnessed the fast growth in telecommunication (Telco) techniques from 2G to upcoming 5G. Precise outdoor localization is important for Telco operators manage, operate and optimize networks. Differing GPS, a technique employed by localize mobile devices using measurement report (MR) data. When given MR samples containing noisy signals (e.g., caused signal interference attenuation), often suffers high errors. To this end, main focus of paper how improve accuracy via algorithms detect repair outlier positions with Specifically, we propose context-aware technique, namely ${\sf RLoc}$ , which consists three components: machine-learning-based algorithm, detection algorithm find flawed samples, replace results better ones (ideally ground truth positions). Unlike most existing works every sample independently, instead take into account spatio-temporal locality locations exploit trajectory context positions. Our experiments on real data sets GSM 4G LTE networks verify that our work can greatly location accuracy. For example, large set achieve 32.2 meters median errors, around 17.4 percent than state-of-the-art.