Precise Point Positioning from Combined GPS, GLONASS & BeiDou Mazher Choudhury Chris Rizos
Challenges of Precise Point Positioning (PPP) GNSS Constellation(s) CORS Network Other models & files GNSS User Precise GNSS Satellite Orbits & Clock Corrections (Real-time or files)
Challenges of Precise Point Positioning (PPP) GNSS Constellation(s) CORS Network Other models & files GNSS User Precise GNSS Satellite Orbits & Clock Corrections
From GPS to Multi-GNSS... + GNSS: GPS (31) (32) GLONASS (24) (30) Galileo (7-8) (30) BeiDou (13-14) (35) RNSS: QZSS (1) (5-7) IRNSS (4) (7) SBAS: WAAS MSAS EGNOS GAGAN SDCM Number of satellites: (Current) (Planned)
More Signals, More Frequencies...
IGS Multi-GNSS Experiment Multi-GNSS Experiment (MGEX) Multi-GNSS Working Group, chaired by Oliver Montenbruck Build-up of new multi-gnss tracking network started 2012 (ongoing) Preliminary MGEX results from variety of ACs http://igs.org/mgex/ About 30 contributing agencies from >16 countries More than 120 stations worldwide, plus 75 real-time stations Tracking of Galileo, BeiDou, QZSS, SBAS signals Free data/product access: ftp://cddis.gsfc.nasa.gov/pub/gps/data/campaign/mgex/ ftp://cddis.gsfc.nasa.gov/pub/gps/products/mgex/
IGS MGEX Network ~120 Stations (Apr. 2015) Offline : ftp://cddis.gsfc.nasa.gov/pub/gps/data/campaign/mgex/ Real-time: http://mgex.igs-ip.net/ 7
IGS MGEX Test Products Currently Galileo, BeiDou, QZSS orbits and clocks 7 contributing ACs: CNES, ESA, CODE, GFZ, JAXA, TUM, WUM Orbits & clocks at decimetre-level accuracy all-in-one constellation broadcast ephemeris file ( brdm ) ISB, DCBs results SLR residuals for several GNSS satellites CODE ESA GFZ GFZ CNES JAXA TUM WHU Status: April 2015
Background to M-GNSS PPP Previous work Cai & Gao (2007): GPS + GLONASS Afifi & El-Rabbany (2014): GPS + Galileo Tegedor et al. (2014): GPS + GLONASS + Galileo + BeiDou (This is implemented using sequential least squares approach within ESA proprietary software NAPEOS) In this work Implemented in SNAPP GNSS data processing software (based on Open Source RTKLIB) Kalman filter is used Extended for real-time PPP & RTK-PPP Galileo measurements are not included
Data Flow Through Software... RINEX multi - GNSS observation file RINEXmulti - GNSS Navigation file Multi-GNSS Precise Orbit and clock file PPP Engine Reading data from file and prechecking for cycle slips, gross errors, etc Initialise states for KF Generate ionospheric-free combinations EOP file Ocean tidal loading Antenna offset file Apply corrections KF processing Estimated parameters
Data Source IGS MGEX provides satellite orbit/clock files for BDS from 28 January 2014 Dual-frequency measurements for GPS Week 1799, DOY 181 (30 June 2014) for two GNSS stations in Australia: GPS, GLO & BDS from MRO1, GPS+BDS from UNX3 Station Receiver Firmware Antenna Tracking GNSS MRO1 TRIMBLE NETR9 4.81 TRM59800.00 NONE GPS+GLO+BDS UNX3 SEPT ASTERX3 2.3.4 LEIAR25.R3 LEIT GPS+GLO
Receiver clock offset(ns) Receiver clock offset(ns) Receiver (GPS) Clock Error Solutions 0 MRO1-5 -10-15 -20 34 32 UNX3-25 GPS Only GPS+GLO+BDS 30-30 86400 104400 122400 140400 158400 GPS time, DOY 181 Multi-GNSS processing has no significant impact on (GPS) Receiver Clock Offset solutions 28 26 24 GPS Only GPS+BDS 22 20 86400 104400 122430 140430 158430 GPS time, DOY 181
ISB(ns) Inter-System Biases Clearly receiver HW dependent... for this dataset 5 MRO1 ISB_GLO MRO1 ISB_BDS UNX3 ISB_BDS 0-5 -10-15 -20-25 -30-35 86400 97230 108030 118830 129630 140430 151230 162030 GPS time, DOY 181
ZTD(m) Zenith Tropospheric Delay Estimates 2.5 2.45 MRO1 UNX3 2.4 2.35 2.3 2.25 2.2 2.15 GPSTime for DOY,181 2.1 86400 97230 108030 118830 129630 140430 151230 162030
M-GNSS PPP Results... Accuracy compared to AusPos solution MRO1 UNX3 E 0.005 (0.009) N -0.002 (0.004) (0.001) H -0.004 (0.011) (0.118) 2D 0.006 (0.010) (0.003) E -0.007 (0.016) N -0.005 (0.008) (0.001) H 0.161 (0.163) (0.113) 2D 0.012 (0.017) GPS GLO BDS GPS+GLO GPS+BDS GPS+GLO+B DS -0.007 0.018 0.002 0.004 0.004 (0.014) (0.047) (0.007) (0.018) (0.008) (0.001) 0.000 (0.007) (0.001) -0.003 (0.013) (0.044) 0.011 (0.015) -0.087 (0.093) -0.300 (0.444) (0.351) 0.097 (0.105) -0.038 (0.085) (0.001) -0.010 (0.035) (0.003) 0.366 (0.421) (0.342) 0.071 (0.092) (0.004) -0.001 (0.003) (0.001) -0.003 (0.010) (0.112) 0.006 (0.008) -0.010 (0.010) (0.001) -0.013 (0.019) (0.073) 0.018 (0.021) -0.003 (0.011) (0.001) -0.007 (0.011) (0.001) 0.169 (0.170) (0.094) 0.011 (0.015) -0.005 (0.006) (0.001) -0.003 (0.014) (0.070) 0.009 (0.010) Row 1: mean Row 2: RMS Row 3: Std Dev
Error(m) Error(m) PPP Results... issue of convergence 0.5 MRO1 E N H 0.5 UNX3 0.4 0.3 GPS+GLO+BDS 0.4 0.3 GPS+BDS 0.2 0.2 0.1 0.1 0-0.1-0.2-0.3-0.4-0.5 86416 100816 115216 129616 144016 158416 0-0.1-0.2-0.3-0.4-0.5 86416 100816 115216 129616 144016 158416 GPS time
Minute Minute PPP Filter/Ambiguity Convergence... Time taken to reach 5cm, 10cm, 20cm, 50cm accuracy 240 220 200 180 160 140 120 100 80 60 40 20 0 MRO1 0.05 0.1 0.2 0.5 E N H E N H E N H E N H E N H E N H GPS GLO BDS GPS+GLO GPS+BDS GPS+GLO+BDS 240 220 200 180 160 140 120 100 80 60 40 20 0 UNX3 0.05 0.1 0.2 0.5 E N H E N H E N H E N H E N H E N H GPS GLO BDS GPS+GLO GPS+BDS GPS+GLO+BDS
Concluding Remarks Kalman filter was implemented for multi-gnss PPP solutions (GPS, BDS and GLO), in which ZTD and ISB were also estimated. Mathematical model & software was tested using 24 hours static data from two IGS stations Sub-decimetre-level accuracy can be achieved within a convergence time of about 40 minutes Future work includes implementation of an adaptive robust Kalman filter, which is expected to improve accuracy and reduce convergence time using knowledge of measurement residuals