Performance of Precise Point Positioning using Current - - PowerPoint PPT Presentation

IGNSS Conference, Sydney, 6 – 8 Dec 2016

Performance of Precise Point Positioning using Current Triple-frequency GPS Measurements in Australia

Viet Duong1, Ken Harima1, Suelynn Choy1, Chris Rizos2

1 School of Science, RMIT University 2 School of Civil and Environmental Engineering, University of New South Wales

Viet Duong | RMIT University 2

Outline

❶ Introduction to triple-frequency PPP ❷ Aims of research ❸ Mathematical models ❹ Methodology ❺ Results & Conclusion

Viet Duong | RMIT University 3

Introduction to Triple-frequency PPP

• NRTK is not suited for precise positioning service
• n a wide-area or global scale
• PPP requires a long convergence time (e.g.

several tens of minutes), not suitable for real-time applications

• Using triple-frequency GPS can speed up

convergence time as shown by Laurichesse (2012, 2015, 2016)

• The data used is limited due to the limited

number of GPS Block IIF satellites and ground infrastructure limitations

Viet Duong | RMIT University 4

Aims at research

• To provide an insight into the current

performance of triple-frequency GPS PPP in Australia

Source: reproduced from Laurichesse (2016)

Number of triple-frequency GNSS satellites in view for a given location

Viet Duong | RMIT University 5

Mathematical models

• Single-differenced-between-satellite method has

been used

• Three main steps for dealing with triple-

frequency PPP:

Viet Duong | RMIT University 6

Steps Wavelengths (m) Noise levels of combinations Parameters Possible advantages for Triple-frequency Dual Triple Dual Triple Dual Triple EWL Not applie d 5.86 Not applied 0.7σP NEWL Ambiguities are FIXED instantaneously WL 0.86 3.40 0.7σP 110σL NWL U+ T+ NWL

• WL ambiguities can be fixed

in about two minutes

• Providing decimetre-level

positioning accuracy NL 0.107 0.108 2.99σP 2.99σL 110σL 2.54σL U+ T+ NNL U+ T+ NNL

• An optimal combination for

phase measurements with a low noise level

• Using the support of WL and

EWL to create a new form “code measurement”

Mathematical models

U: user positions; T: troposphere delays; N: Ambiguities

Viet Duong | RMIT University 7

Methodology – Static test

• The research used real observations collected

from eight Australian CORS stations over one week period in 2016

• Triple- and dual- frequency PPP results were

compared for both float and fixed ambiguities

• The proposed algorithms in Laurichesse (2012,

2015, 2016), Li et al. (2013), Geng and Bock (2013) have been implemented

• A modified version of RTKLIB and Matlab-based

GPS PPP data processing software have been developed.

Viet Duong | RMIT University 8

Methodology – Static test

Visibility of the GPS Block IIF satellites transmitting L5 signal at the ALIC station on DOY 215 2016 8 GNSS CORS stations from the Australian Regional GNSS Network (Source: Google Earth)

Viet Duong | RMIT University 9

10 20 30 40 50 60 70 80 90 0.1 0.2 0.3 0.4 0.5

Horizontal RMS(m)

RMS-Hz-Dual RMS-Hz-Triple+Dual

10 20 30 40 50 60 70 80 90

Time (minutes)

0.1 0.2 0.3 0.4 0.5

Vertical RMS (m)

RMS-Up-Dual RMS-Up-Triple+Dual

10 cm: 2F&3F ~ 22 min

Results

• Float solution for dual- and triple- PPP

approaches

Viet Duong | RMIT University 10

10 20 30 40 50 60 70 80 90 0.1 0.2 0.3 0.4 0.5

Horizontal RMS (m)

RMS-Hz-Dual RMS-Hz-Triple+Dual

10 20 30 40 50 60 70 80 90

Time (minutes)

0.1 0.2 0.3 0.4 0.5

Vertical RMS (m)

RMS-Up-Dual RMS-Up-Triple+Dual

10 cm 3F : ~ 9 2F : ~ 14min

Results

• Fixed solution for dual- and triple- PPP

approaches

Viet Duong | RMIT University 11

Results

• Ambiguity fixing rate

10 20 30 40 50 60 70 80 90

Time (minutes)

10 20 30 40 50 60 70 80 90 100

Ambiguity Fixing Rate (%)

SR-Dual SR-Triple+Dual

AFR > 95%, after 20min

Viet Duong | RMIT University 12

Horizontal RMS threshold Convergence time (minutes) Vertical RMS threshold Convergence time (minutes) Dual float Dual fixed Triple float Triple fixed Dual float Dual fixed Triple float Triple fixed < 5cm N/A 22 N/A 17 < 5cm N/A 26 N/A 17 < 10cm 25 15 23 10 < 10cm 24 12 24 10 < 15cm 15 9 9 7 < 15cm 12 10 11 7

Results

• Positioning convergence time with combined

results from eight selected stations

*N/A: Data collected is not enough to reach accuracy levels.

Viet Duong | RMIT University 13

Conclusion

• Dual- and triple- frequency PPP GPS

performance in Australia has been demonstrated

• Triple-frequency GPS satellites:

– strengthen the PPP positioning model, – improve the float ambiguity positioning accuracy & shorten solution convergence time by an average of 5 minutes compared to dual-frequency PPP – Increase the ambiguity fixing success rate by 10%

• Future work: Adding multi-constellation and

multi-frequency GNSS to the PPP model, expanding with kinematic mode

Viet Duong | RMIT University 14

Viet Duong | RMIT University 15

References

• Laurichesse, D 2012, Phase Biases Estimation for Undifferenced Ambiguity Resolution,

paper presented to PPP-RTK & Open Standards Symposium, Frankfurt, Germany.

• Laurichesse, D 2015, Handling the Biases for Improved Triple-Frequency PPP

Convergence, GPSWorld.

• Laurichesse, D 2016, Fast PPP Convergence Using Multi-Constellations And Triple-

Frequency Ambiguity Resolution, paper presented to IGS, Australia.

• Li, T, Wang, J & Laurichesse, D 2013, Modeling and quality control for reliable precise

point positioning integer ambiguity resolution with GNSS modernization, GPS Solutions, vol. 18, no. 3, pp. 429-42.

• Geng, JH & Bock, Y 2013, Triple-frequency GPS precise point positioning with rapid

ambiguity resolution, Journal of Geodesy, vol. 87, no. 5, pp. 449-60.

• El-Mowafy, A, Deo, M & Rizos, C 2016, On biases in precise point positioning with

multi-constellation and multi-frequency GNSS data, Measurement Science and Technology, vol. 27, no. 3, p. 035102.