Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
THE GPS BROADCAST ORBITS:AN ACCURACY ANALYSIS
R.B. Langley1, H. Jannasch1, B. Peeters1,2, and S. Bisnath1
Session B2.1-PSD1, New Trends in Space Geodesy33rd COSPAR Scientific Assembly, Warsaw, 16-23 July 2000
1Geodetic Research LaboratoryUniversity of New Brunswick
Fredericton, N.B., Canada
2Faculty of Aerospace EngineeringDelft University of Technology
Delft, The Netherlands
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Outline
• Motivation
• Goals
• Approach
• Results
• Web Display
• U.S. Space Command Reports
• Anticipated Broadcast Orbit Improvements
• Conclusions and Future Work
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Motivation
• Day-to-day accuracy of navigation message ephemerides(broadcast orbits) not readily available• Independent assessment of ephemeris error in StandardPositioning Service error budget considered useful• Broadcast orbit accuracies needed to assess errors infuture U.S. and Canadian Wide Area AugmentationSystems (WAASs)
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Goals
• Develop algorithms and software to automaticallycompute (once per day) the broadcast orbit errors andmeaningful statistics
• Develop algorithms and software to automaticallycompute (once per day) the WAAS orbit correction errorsand meaningful statistics
• Post the analysis results on the World Wide Web
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Approach
• Accuracies are assessed through comparisons with orbitproducts provided by the International GPS Service (IGS)
• Separate assessments carried out with IGS predicted,rapid, and final orbits
• Broadcast orbits obtained from “auto” RINEX navigationfiles provided by Scripps Orbit and Permanent ArrayCenter (SOPAC) Web site
• IGS orbits obtained from the IGS Central Bureau ftp site
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Approach, cont’d.
• Broadcast orbits evaluated according to ICD-GPS-200
• Ephemeris sets are normally updated every 2 hours
• Ephemeris set considered valid from 2 hours beforeTime of Ephemeris (TOE) until 2 hours after TOE
• But a particular set is first used when it was firsttransmitted = Transmission Time of Message (TTOM),typically 1:59:42 before TOE
2:00:00 8:00:00
4:00:001:59:44
0:00:00 2:00:00 4:00:00 6:00:00 8:00:002 5
10 11
TimeIODE
TOCIODE
TOC 0:00:001
Normal sequence of Ephemeris sets
Updates on the Ephemeris
IODE = Issue of Data Ephemeris
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Approach, cont’d.
• Antenna phase centre offset from spacecraft centre ofmass:
Block II/IIA*: xpc = 0.279, ypc = 0.000, zpc = 1.023 metresBlock IIR*: xpc = 0.279, ypc = 0.000, zpc = 1.023 metres
• Standard body-fixed frame to WGS 84 conversionapplied
• No datum transformation; WGS 84 assumed consistentwith IGS frame (ITRF)
*NIMA usesBlock II/IIA*: xpc = 0.2794, ypc = 0.0000, zpc = 0.9519 metresBlock IIR*: xpc = 0.0000, ypc = 0.0000, zpc = 1.1725 metres
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Approach, cont’d.
• Broadcast orbits computed every 15 minutes at the IGSorbit epochs
• Differences between broadcast orbit x, y, z values andIGS values computed = dx, dy, dz
• For each day, the minimum, maximum, and r.m.s.differences, dx, dy, dz for each satellite are computed
• 3D-error: also computed
• Results posted to Web
dx 2 + dy 2 + dz2
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Constellation Daily 3D R.M.S. Error
Broadcast orbit R.M.S. error with respect to IGS final orbit(constellation 3D R.M.S.)
0
5
10
15
20
25
0 50 100 150 200 250 300 350 400 450 500
Day (January 1, 1999 - June 5, 2000)
Constellation Overall R.M.S. = 5.03 mPRN6 ∆V
PRN18 ∆V
PRN 15 ∆V
PRN26 ∆V
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Some Problem Satellites
• PRN02 - possible problems in eclipse
• PRN14 - large ranging errors; retired 14/4/00
• PRN15 - reaction wheel problems in eclipse
• PRN16 - clock problems
• PRN18 - reaction wheel problems in eclipse
• PRN23 - solar array slewing in eclipse
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN14 Daily 3D R.M.S. Error
PRN 14 broadcast orbit R.M.S. error with respect to IGS final orbit(3D R.M.S.)
0
5
10
15
20
25
1 51 101 151 201 251 301 351 401 451 501
Day (January 1, 1999 - June 5, 2000)
PRN14 in eclipse
PRN14 3D RMSOverall R.M.S. = 7.71 m
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN15 Daily 3D R.M.S. Error
PRN 15 broadcast orbit R.M.S. error with respect to IGS final orbit(3D R.M.S.)
0
5
10
15
20
25
1 51 101 151 201 251 301 351 401 451 501
Day (January 1, 1999 - June 5, 2000)
PRN15 in eclipse
PRN15 3D RMS
Overall R.M.S. = 7.25 m
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN18 Daily 3D R.M.S. Error
PRN 18 broadcast orbit R.M.S. error with respect to IGS final orbit(3D R.M.S.)
0
5
10
15
20
25
1 51 101 151 201 251 301 351 401 451 501
Day (January 1, 1999 - June 5, 2000)
PRN18 in eclipse
PRN18 3D RMSOverall R.M.S. = 7.94 m
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN13 Daily 3D R.M.S. Error
PRN 13 broadcast orbit R.M.S. error with respect to IGS final orbit(3D R.M.S.)
0
5
10
15
20
25
1 51 101 151 201 251 301 351 401 451 501
Day (January 1, 1999 - June 5, 2000)
PRN13 in eclipse
PRN13 3D R.M.S.Overall R.M.S. = 3.18 m
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
3D R.M.S. Error by Satellite
Broadcast orbit R.M.S. error with respect to IGS final orbit(January 1, 1999 - June 5, 2000)
0
2
4
6
8
10
1 2 3 4 5 6 7 8 9 10 11 13 14 15 16 17 18 19 21 22 23 24 25 26 27 29 30 31
Satellite PRN number
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
3D R.M.S. Error on March 8, 2000
Broadcast orbit R.M.S. error with respect to IGS final orbit(March 8, 2000)
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7 8 9 10 11 13 14 15 17 18 19 21 22 23 24 25 26 27 29 30 31
Satellite PRN number
Overall R.M.S. of day = 7.36 m
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
3D R.M.S. Error on May 14, 2000
Broadcast orbit R.M.S. error with respect to IGS final orbit(May 14, 2000)
0
2
4
6
8
10
12
14
16
18
20
1 2 3 4 5 6 7 8 9 10 11 13 15 16 17 18 19 21 22 23 24 25 26 27 29 30 31
Satellite PRN number
Overall R.M.S. of day = 3.60 m
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN13 XYZ Errors on March 8, 2000
PRN 13 - Broadcast orbit error with respect to IGS final orbit
-3
-2
-1
0
1
2
3
1 11 21 31 41 51 61 71 81 91
Epoch of day (March 8, 2000)
dX
dY
dZ
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN14 XYZ Errors on March 8, 2000
PRN 14 - Broadcast orbit error with respect to IGS final orbit
-15
-10
-5
0
5
10
15
20
1 11 21 31 41 51 61 71 81 91
Epoch of day (March 8, 2000)
dX
dY
dZ
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN13 RTN Errors on March 8, 2000
PRN 13 - Transformed broadcast orbit error with respect to IGS final orbit
-2
-1
0
1
2
3
4
1 11 21 31 41 51 61 71 81 91
Epoch of day (March 8, 2000)
Radial
Transverse
Normal
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
PRN14 RTN Errors on March 8, 2000
PRN 14 - Transformed broadcast orbit error with respect to IGS final orbit
-5
0
5
10
15
20
25
1 11 21 31 41 51 61 71 81 91
Epoch of day (March 8, 2000)
Radial
Transverse
Normal
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Web Display
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Web Display cont’d.
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Web Display cont’d.
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Satellite Incidents
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Signal-in-Space Range Error
• Computed by GPS Control Segment
• Clock and ephemeris errors:
whereR = radial ephemeris errorA = along-track ephemeris errorC = cross-track ephemeris errorCLK = clock error
SISRE = R − CLK( )2 +1
49A2 + C2( )
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Signal-in-Space Range Error Estimates
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Anticipated Broadcast Orbit Improvements
• Broadcast orbits should improve due to the GPS AccuracyImprovement Initiative (AII) and other efforts:– augment tracking network with NIMA stations
– improve Master Control Station Kalman filter
– more frequent uploads
– expected r.m.s. SISRE after AII = 1.3 metres
– further enhancements from clock replacements at monitor stations,multipath mitigation, improved tropospheric delay modelling,more satellites with rubidium clocks
Geodetic Research Laboratory • Department of Geodesy and Geomatics Engineering • University of New Brunswick
Conclusions and Future Work
• Daily broadcast orbit error automatically computed andposted to the Web
• Database available for detailed study
• Constellation 3D r.m.s. error over past 18 months about 5metres;some satellites as good as 3 metres
• RTN error components to be computed and compared toMCS SISRE estimates
• Comparison with NIMA precise ephemeris
• WAAS orbit correction errors to be assessed