A SON enhanced algorithm for observed time differences based geolocation in real 3G networks

This paper presents a SON enhanced approach for OTD Based Geolocation in a real 3G UMTS Network. The algorithm was implemented using cell, measurement report and drive-test data, the last just for geolocation accuracy determination. It uses a dual convergence approach evolving Node B time reference estimation followed by trilateration, boosting convergence. A motion model Kalman filtering module was developed, and tuned for urban mobile. Around 35% of the events were geopositioned, which motivates using OTD measurements for geolocation in live networks. The median accuracy achieved by the algorithm was 178 m, which is rather promising, considering that OTD measurements have an intrinsic uncertainty of a UMTS chip, i.e, 78 m.