Improving the map matching on passive mobile positioning data by minimizing the effects of cell geometry data

This paper addresses cell data quality problems encountered in map matching based on passive mobile positioning data. The passive positioning data is typically on cell global identifier level, i.e. data record tells that the mobile station was connected to given cell at given time. Each cell has attributes that can be used to estimate the effective area of the cell. The assumed cell area can very significantly differ from the reality which would result in very unrealistic map matching results. The paper describes a technique to find conflicting cell data and proposes an algorithm for utilizing this knowledge so that conflicting cell data is suppressed in map matching process. The technique significantly improves the map matching results. Keywords—mobile positioning; cellplan; data quality; map matching