Correction Of Telecom Localization Errors By Context Knowledge

Telecom localization that had aroused widespread attentions of major telecommunication operators has become vital in recent years. However, current available technologies suffer from high localization errors, typically with mean errors more than 100 m. In order to tackle this problem, in this paper we leverage context knowledge to reduce the localization error. To this end, we propose a framework adopting several modified filter methods in terms of context to eliminate localization errors that cannot be easily detected by the existing localization algorithms. We apply the optimized filter methods combining with the context knowledge to verify the effectiveness of our methodologies according to the experiments based on the telecom localization utilizing the GPS-associated MR data in the downtown area of Shanghai, China.