Table 11 summarises the error budget for GPS in the marine
navigation context. This implies typically single frequency C/A code receivers under the
influence of SA.
Table 1: Error Budget of GPS observations including effects of SA
Error Contributor
Remarks
Amount(Meteres)
Clock Error
Principally due to the error of the Satellite clock.
3.0m
Ephemeris Error
Resulting error of position of the Satellite due to quality of the provided data
2.4m
S.A Dither
Imposed error on the Ephemeris data to dilute accuracy to SPS levels.
24.0m
SA. Epsilon
Imposed error on the Satellite clock to dilute accuracy to SPS levels
24.0m
Ionospheric Errors
Residual error after model derived corrections have been applied
4.0m
Troposphereic Errors
These are typically relatively small.
0.4 m
Receiver Errors
Emperical estimate
1.0m
The above gives a rms value of 34.4 m and a drms value of 103.4 metres,
which closely approximates the SPS specification.
Some Solutions
In order to obtain more accuracy and consistency from GPS observations,
various strategies have been devised, largely centred on the extensive correlating effect
of SA dither and epsilon effects, which vary almost linearly with range from the monitor
station. These are briefly described below:
Conventional DGPS
Convention Differential GPS (CDGPS) is a mechanism to compensate
correlative effects obtained by analysis of monitor station observations and remote
receiver observations. The fundamental elements needed are the simultaneous measurement of
observations at the remote receiver and the monitor station. A simple application for a
casual experiment is to have a mobile telephone link and ensure that monitor and remote
receivers are both using the same satellites to derive position, without the intrusion of
filters in the computation process. In such a case, the difference between the known
coordinates of the monitor station when subtracted from its observed position will give
latitude and longitude corrections to apply to the remote receiver to obtain its true
position. High accuracies can be obtained in this manner but this method has several
disadvantages.
The first is that both receivers must have the same satellite
configuration. The second is that this has limited value other than in post processing of
the remote receiver observations. DGPS systems therefore invariably transmit corrections
in the MF band, to be applied to the pseudo ranges obtained from the Code Phase
observations on the L1 frequency. The GPS receiver typically obtains these over a standard
serial link In NMEA format from a medium frequency radio receiver. The GPS receiver then
applies these corrections to the observed Code Phase pseudo ranges before calculating
position. This position is typically of the order of 3.5 to 8.9 metres over the 500 nm
range. at 2 drms levels with an HDOP of 1.52 While large numbers of these
stations have been installed all over NW Europe and the United States coastal confluence,
it is possible that the position of SA will be reviewed before the year 2000. A decision
to revoke SA may well dilute the importance of this Investment for marine navigation due
to the need for fairly low levels of accuracy, which may be satisfied by the 30 meter
accuracy3 that SPS would provide in the absence of SA.
