With classical real time kinematic surveying (RTK) using GPS, the correction data of one reference station are transmitted to the user. However, the distance between the reference station and the mobile user (rover) should not exceed 15-20 km, otherwise the measurements will be notably affected by the systematic errors. If one wished to develop a network of GPS permanent stations over an entire country, one would need a very dense and in effect also a very expensive network.
The concept of virtual reference stations (VRS) offers new possibilities. The principle is to interpolate the data of several reference stations in order to obtain the correction data for the rovers, which reduces the systematic influences of the RTK measurement decisively. Not only may the allowed distance between the reference station and the rover be increased, but also the reliability of the system is heightened. Should a reference station fail temporarily for example, the correction data are computed with the surrounding reference stations. In addition, the productivity is improved by clearly shorter initialization times.
Virtual reference station
A virtual reference station is an imaginary, unoccupied reference station which is only a few meters from the RTK user. For this position, observation data are created from the data of surrounding reference stations as though they had been observed on that position by a GPS receiver.
Principle
- Operating mode of the virtual reference station
Procedure
- The data from the AGNES stations are transmitted to the processing and control center at a one-second interval
- The RTK user sends his approximate position (NMEA string)
- The processing center determines in which triangle the user is stationed and computes a virtual reference station (as a function of the transmitted approximate position)
- The processing center transmits the data of the virtual reference station to the user (RTCM format)
GPS error
Using the concept of the virtual reference stations, certain systematic errors can be reduced considerably:
- Ionosphere
- Troposphere
- satellite orbits
- Multipath
- Antenna phase center
One of the main sources of errors is the Ionosphere . The electrons in the ionosphere are ionized by solar radiation which leads to errors up to 50 m in the GPS observations. The variation of the electron density in the ionosphere depends on the solar cycle which lasts 11 years. The influence increases with the number of solar spots (a maximum was recorded in 2002) thus making efficient modelling indispensable.
The influence of the Troposphere depends on the weather conditions (pressure, temperature, atmospheric humidity , etc). This influence can be corrected with the appropriate models.
The influence of the satellite orbits can also be corrected using GPS ephemerides.
Multipathing is caused by the direct or indirect reflection of the GPS signals, which distorts positioning. Especially critical are metal surfaces in the vicinity of GPS antennas. It is difficult to correct multipathing using models.
The Antenna phase center is an approximate calibration value indicated by the software firms. The actual phase center, however, may deviate from these values by several millimeters. In order to obtain precise antenna phase centers, each antenna must be calibrated separately.
Software
The positioning service swipos-GIS/GEO uses the network software "GPS-Net" from Trimble Terrasat GmbH. GPS-Net is based on the principle of the virtual reference station. Of the different types of errors mentioned above, special emphasis is placed on the correction of the ionosphere and the satellite orbits. The systematic errors are divided into two groups:
- geometric corrections (troposphere and ephemeride errors)
- ionospheric corrections