Title: GNSS data products, post processing and real time services and their applications in the context of
1GNSS data products, post processing and real time
services and their applications in the context
of AFREF
- AFREF Technical Workshop
- 10-13 July 2006
- UCT
- Ludwig Combrinck
2AFREF Products
- Products that do not necessarily require real
time data - Coordinates (the MAIN product)
- Coordinate velocities
- Meta information
- Products that do require real time data
- Differential correction services (DGPS and RTK)
perhaps an AFRICA DGPS based on cell phone
/internet technology - Total electron content maps
- Integrated water vapour maps for operational
weather forecasting - For crustal motion (Earthquakes, volcano
eruptions) - Early warning systems (e.g. Indian Ocean Tsunami
Early Warning System)
3How to get to products
- Need to define
- a starting epoch for the AFREF datum (e.g 1
January 2008) and the AFREF specific geodetic
datum (the set of parameters that describe the
size and orientation of the earth etc.) - a set of core AFREF stations (the 1st order
network) - a processing strategy to obtain ITRF coordinates
for the core stations - A procedure/centre to maintain the core AFREF
stations (positions, velocities, data integrity,
metafiles etc.) - A set of guidelines for users of AFREF to enable
them to use the products efficiently and
accurately (AFREF cookbook)
4And then
- Strategies within the various SG depts to utilise
the AFREF products - Campaign surveys
- AFREF densification (2nd order network)
- Network adjustments/transformations
5Baseline Solutions/Network Position Determinations
- 1st order network
- A processing strategy needs to be decided upon
before processing data - For instance, a selection of global IGS core
stations must be selected and included in a
global solution to compute AFREF ITRF positions
and - careful thought must be given to the weighting of
stations - 2nd order network
- GPS baseline solutions are generated through an
iterative process. - From approximate values of the positions occupied
and observation data, theoretical values for the
observation period are developed. - Observed values are compared to computed values,
and an improved set of positions occupied is
obtained using least-squares minimisation
procedures and equations modeling potential error
sources. - Observed baseline data are also evaluated over a
loop or network of baselines to ascertain the
reliability of the individual baselines.
6Typical flow of process to measure baselines from
AFREF core station/s to local new Datum point
- Download AFREF core station data precise
ephemeris logfile from AFREF data centre - Download baseline data from receivers (fixed
stations and campaign data) - Download and prepare all other necessary input
data such as station heights, antenna models - Make changes and edit raw baseline data (e,g,
reject short observational periods etc) - Use AFREF core station as fixed with its
position as at AFREF Datum epoch and process all
baselines - Review, inspect, and evaluate adequacy of
baseline reduction results - Make changes and reject bad baselines
- Reprocess baselines and re-evaluate results
- Note/Designate independent and trivial baselines
- Review closures and adjust baseline network
- End Product a set of ITRF coordinates relative
to AFREF Datum (say 1 January 2008).
7How to do the processing (or with what)
- On relatively short baselines, up to several
hundreds of kilometres, commercial, off the shelf
software (Trimble, Ashtech etc.) will probably be
fine - However, for longer baselines, say 500 km , one
should rather use software that adequately
incorporates earth tide modelling and other
effects on position - It will depend on your processing and network
occupation/reoccupation strategy (e.g.
selfcentring plates on trig beacons) and levels
of accuracy required of your new network
8Unless GIPSY (JPL software) is the processing
software, carrier phase differencing will be
used (GAMIT etc.)
The results of the Triple Difference baseline
solution can then be input back into the Double
Difference equations in order to resolve, or
"fix," the integers in the Double Difference
solution. Fixing the integers in a Double
Difference solution constrains the integer
ambiguity N to a whole number of cycles, and
is the preferred baseline solution
9Commercial Software (Trimble, Ashtech etc.)
- Baseline processing software is now fairly
automatic and user-friendly. - Allows a processing strategy to be set up
- Most software automatically performs all the
differencing operations needed to solve for
integer ambiguities - Displays the resultant baseline vectors along
with adjustment and accuracy - Generate reports and statistics that can be used
to evaluate the results.
10Non-commercial scientific software (including
Bernese)
- BERNESE, developed by the University of Bern,
GAMIT developed by MIT, GEODYN by NASA, GIPSY by
JPL, TOPAS, Germany, GPSOBS by ESA, TEXGAP by
Univ. Texas, GEPHARD by GFZ, Germany, etc. - Are not very user friendly
- Could need advanced knowledge of operating
system, Fortran compiler etc. - Mostly need knowledge of C-shell scripts
- Good background in processing strategies and most
of the time, experience - So, bottom line for this is, a steep learning
curve and time invested, but is worthwhile
11Solution Acceptance Criteria
Evaluation of results is important, as they
vary and you may have to query (or reject)
measurements
12Residual Plots typically the L1 phase residual
error is plotted
Residual plots depict the data quality of the
individual satellite signals, typically vary
around 5 mm from the mean. Deviations exceeding
15 mm are suspect. Multipath problems may show
up on the residual plot as a sinusoidal wave
over time and is best minimised by good site
selection, choke ring antennas, and long session
times which allow a change in satellite geometry.
13Do it yourself or
- We could look towards establishing a processing
centre for AFREF which would take your data,
process it and give you the AFREF coordinates - It depends on what is realistically achievable in
each country within the time frame you have set
yourself
14Real time products
Could be added to global WADGPS systems such as
Starfire (accuracy about 35 cm horizontal)
15Starfire reference network
Source NavCom Technology Inc.
16Could be used for
17and share the vision..
18Africa sparse contribution at this time
19SourceStreaming Real-Time IGS Data and Products
Using NTRIP, Georg Weber IGS Workshop, May 08-11,
2005, Darmstadt, Germany
20In terms of AFREF
- The sky is the limit
- But we have to get beyond the talking point. and
start doing something. soon!
The End