Learning Center
Plans & pricing Sign in
Sign Out
Get this document free



									Presentation type (select one):
   Oral (Invited only)
   Oral Request (Vertical Rates Workshop)
   Poster (Vertical Rates Workshop)

Title Text
  Introduction to PANDA Software and the latest Development for high precision
                     GNSS Data Processing and Application
  Chuang Shi1, Maorong Ge2, Qile, Zhao1, Yidong, Lou1, Jianghui Geng3, Min, Li1, Jingnan
                                 Liu1, Hongping Zhang1
(1) Wuhan University, P.R. China; (2) GFZ Potsdam, Telegrafenberg A17, Germany; (3)
Nottingham University, U.K.
PANDA (the abbreviation of Positioning And Navigation Data Analyst) is a comprehensive software
designed for high precision GNSS Data Processing and Application. This paper mainly introduces
the functional and algorithms designed in PANDA. Newly results are showing that PANDA can
achieve the same performance as GAMIT, Bernese and GIPSY proved. The mainly functional
designs include:
GNSS SV POD is the basic functional model in PANDA. Now, the precision of GNSS SV POD can
achieve at the level of 3-4cm comparing with IGS final products.
2. Integrated adjustment of GPS and LEO data for Precision Orbit Determination (POD).
The improvement of LEO orbits with the integrated adjustment using PANDA is presented. When 43
ground stations are used, the improvement can get to 17% and 32% without accelerometer data for
CHAMP and GRACE, respectively. So it is demonstrated that the LEO orbit accuracy is improved
simultaneously with the GPS orbits in the integrated adjustment.
3. POD of LEOs and Gravity field modeling
Procedures for both reduced-dynamic and kinematic orbit determination are used in LEO POD and
gravity recovery. The procedure makes use of zero-differenced measurements. A deliberately
developed method for fixing ambiguities of GPS phase measurements acquired by a satellite
formation is presented. The accuracy of 100 days’ orbits is assessed by comparing them with
DEOS, JPL and TUM reduced-dynamic orbits as well as with SLR data and KBR data. The accuracy
of the position is about 2-3 cm while the baseline of two GRACE satellites is about 2-3mm for
dynamic orbits and 4-5mm for kinematic ones.
We have computed two sets of monthly gravity field solutions using the data from January 2005 to
December 2006 with PANDA software. The main difference in the production of our two types of
model is how to deal with nuisance parameters. One way is to remove the nuisance parameters
beforehand. The other way is to estimate the nuisance parameters and geopotential coefficients
simultaneously. The discrepancies of the produced gravity field models are mainly in the range from
degree 2 to 5. This can be explained by the fact that the nuisance parameters and the gravity field
coefficients are highly correlated, particularly at low degrees.
4. PPP
Currently, integer ambiguity resolution at a single station has been performed for a global network.
Daily positioning accuracy can be improved from 3.5 mm, 2.3 mm and 6.3 mm to 1.9 mm, 2.1 mm
and 5.8 mm for the East, North and Up components. Correspondingly, for hourly data, positioning
accuracy can be improved from 3.8, 1.5, 2.8 cm to 0.5, 0.5, 1.4 cm, and real-time ambiguity fixing
can significantly improve the positioning accuracy from 13.7, 7.1 and 11.4 cm to 0.8, 0.9 and 2.5 cm.
Furthermore, the re-convergence periods to ambiguity-fixed solutions are reduced on average from
around 1000 s to roughly 1-2 s with a success rate of over 98% of all re-convergence cases.
5. Real-Time PANDA
The real-time orbit and clock offset determination for GPS satellite can be done by Real-Time
PANDA. The Real-Time GPS orbit has achieved the accuracy of 5cm, equal to the accuracy level of
IGS rapid precise orbit products. The real time clock offsets have been estimated epoch wise, and
the accuracy is better than 0.2ns compared with the IGS finial clock offset products. Using real time
GPS orbits and SV clocks, we build a demo system for the high precision real-time positioning at the
level of centimeters for both single- and dual-frequency GPS terminals.
6. Global Ionospheric Maps (GIM)
 Surface Spheric Harmonic Function (SHF) is used to modeling the GIM, which is the same
algorithm as CODE. But the GIM Modeling is done in a new coordinates system in which the real
sun-earth direction is fixed as Z-axis with DE405 involved to determine the direction. X-axis and Y-
axis is just determined through rotating the ECEF Z-axis to our new Z-axis. The ionospheric Pierce
Points (IPPs) from IGS GNSS data (both GPS and GLONASS) show that there are still several
holes in its global distribution. So, the Grid VTEC derived from our SHF coefficients has some
negative results although the DCB of both GNSS Satellites and receivers are almost consist with
IGS products. The weight strategies and data cleaning up for GIM according to the IPP’s global
distribution are still in tuning for our better GIM products.

To top