Quality of virtual data generated from the GNSS reference station network
Hakli, Pasi
VRS (Virtual Reference Station) approach is based on the virtual data and it is the basis for all computation at the user end. Since the quality of virtual data is the backbone to the results, FGI decided to investigate the quality of virtual data. Study was implemented by computing zero-baselines between virtual and reall GPS data at the locations of stations of the Finnish permanent GPS network, FinnReff. Both long-term and short-term variations of the virtual data were studied. Long-term quality was studied by computing time series of daily static solutions. Computed time series show that the quality of virtual data is quite homogeneous. With daily static solutions standard deviations are in the range of few millimetres for all stations and coordinate components. From daily time series we can also conclude that the accuracy is highly dependent on the coordinates of the VRS network since a clear change in trends was seen after the update of the coordinates. Short-term quality was studied by computing short static and kinematic sessions with virtual data. Hourly and kinematic time series show more details than daily solutions and some daily periods are seen. Periodicity may indicate either to remaining biases in the network parameter estimation or unsuccessful baseline processing. Latter may be caused by poor satellite constellation or bad data quality. Kinematic solutions give a picture of the accuracy that could be expected in real-time surveys.
Event: XXIII International FIG Congress : Shaping the change
Only personal, non-commercial use of this document is allowed.