幻灯片 1

Size: px

Start display at page:

Download "幻灯片 1"

赈引祖
5 years ago
Views:

1 Realization of Terrestrial Reference Frame for GNSS Zhang, Hongping on behalf of Jingnan Liu, Chuang Shi, Rong Zou, Na Wei GNSS Research Center Wuhan University, Hubei, China ICG WG-D, Oct

2 OUTLINE Back Ground TRF Realization by Multi-techniques combination A-optimal regularization Intra-technique combination Inter-technique combination Compass Reference Frame simulation Geocenter motion from GPS 2

3 BackGround TRF is a basic for GNSS, each system has to develop its TRF TRF is complicate in definition and realization TRF needs data and software support from global or regional area TRF Realization is an important step for GNSS TRF has to meet all kinds of application in deferent levels of precision 3

4 BackGround ITRF

5 What s TRF TRF Reference Frame: Points with accurate xyz Inputs: Points with X Y Z, freedom (observed by GPS,VLBI,SLR) aprior information as EOP, some sites coordinates Outputs: Points with XYZ and velocity field Transformation Parameters, EOP Purpose: determine how these points are consistent with each other, so that they are in a family Problem in Realization of TRF? points with freedom, Rank deficiency 5

6 Datum Estimation Minimum constriants ( )ˆ 2 T 2 T N+ k B B x= b+ k B c Over-constrained GPS NEQ??? A-optimal regularization ( N + λ I) xˆ = b Fig. 1 Condition number of 78 GPS SINEX files 6

7 Regularization parameter Fig. 3 Statistical Fig. 2 Condition quantities number variations variations wrt regularization wrt k 2 parameters 7

8 Results & Analysis Fig. 4 Differences when k 2 =3*10 7, λ = X Y Z <5mm 22.51% 9.76% 15.23% 5-10mm 72.38% 21.26% 45.99% 10-15mm 5.09% 48.62% 22.16% 15-20mm 0.02% 18.79% 15.62% 20-30mm 0.01% 1.53% 0.98% >30mm 0.00% 0.05% 0.02% Coordinate differences between two individual datum constrains are within 3cm, and 98% of them less than 2cm 8

9 Combination Strategy 9

Intra-technique SLR VLBI GPS Combination combination AC Data-span Station SINEX Constraints Source BKG ASI AC 1993-2009 Data-span Station 70 18 SINEX 2.10 2.

00 Loose loose unconstrained constraints cddis NRC/EMR GSFC GFZ 1993-2009 1999-2009 81 11 71 1.00/2.01 2.

10 Intra-technique SLR VLBI GPS Combination combination AC Data-span Station SINEX Constraints Source BKG ASI AC Data-span Station SINEX loose Loose Constraints constrained constraint Source cddis DGFI COD / Loose loose unconstrained constraints cddis NRC/EMR GSFC GFZ / minimum loose Loose constrained constraints cddis JCET IAA ESA / Loose loose unconstrained constraints cddis USNO NSGFZ / loose Loose constrained constraints cddis JPL /2.01 loose constraints cddis MIT /2.01 loose constraints cddis NCL /2.01 loose constraints cddis NGS /2.01 significant constraints cddis SIO /2.01 loose constraints cddis 10

11 Inter-technique Combination TC Data-span Sta. Num. SINEX Constraints Solution Source IGS minimum constraints Variance-covariance cddis IVS unconstrained Normal equation cddis ILRS unconstrained Variance-covariance cddis 11

12 TRF Simulation Test Coordinate System Defintion ITRS definition IERS 2003 NetWork Global distributed permanent stations Simulation data AC SINEX files Data-span AIUB 28 Weeks ESA 28 Weeks GFZ 28 Weeks Software PowerADJ- PANDA Remarks: GPS weeks from 1399 to 1402, 1419 to 1422, 1431 to 1434, 1443 to 1446, 1460 to 1463, 1471to 1474, 1483 to

13 Results Fig. 5 Transformation parameters of each AC wrt intra-technique combination(whu) 13

14 Analysis good consistency wrt ITRF2005 Fig. 6 Transformation parameters of WHU wrt ITRF

15 Why Geocenter Motion? CM CF X CM V () t = x() tdm V dm X CF () t = Ω x() tdm Ω dm Ideal realization of CF Tt ( ) = 0 0 Δ T = 0 Origin of ITRF Tt ( 0) = 0 T& = 0 From Lavallee,2006 Δ T = No perfect geophysical models available to constrain ΔT : long-term, seasonal, residuals T & = 0 is long-term constraint, and in this case geocenter motion mainly shows seasonal variations 0 15

Degree-one deformation approach Data and preprocessing IGS reprocessed weekly SINEX : 2000.0~2010.0 Network: 132 reference frame stations of IGS05 Linear velocity and jumps : IGS05_repro.

16 Degree-one deformation approach Data and preprocessing IGS reprocessed weekly SINEX : ~ Network: 132 reference frame stations of IGS05 Linear velocity and jumps : IGS05_repro.snx Schemes Sch. 1: Translation Sch. 2: Translation + degree-two mass load Sch. 3: Translation + Rotation Sch. 4: Translation + Rotation + degree-two mass load 16

17 Results 20 X(mm) Y(mm) Z(mm) year Fig. 7 Geocenter motion time series from scheme 2 17

18 Annual terms analysis x y z Amplitude,mm Phase,deg Amplitude,mm Phase,deg Amplitude,mm Phase,deg scheme ± ± ± ± ± ± 1 scheme ± ± ± ± ± ± 1 scheme ± ± ± ± ± ± 1 scheme ± ± ± ± ± ± 1 scheme 2 b 3.92 ± ± ± ± ± ± 1 ig ± ± ± ± ± ± 8 Lavallee,2006 c 3.57 ± ± ± ± ± ± 1 Dong,2003 d 2.1 ± ± ± ± ± ± 3 SLR e a Amplitude defined as A*sin(omega*(t-t 0 )+phase), where t0= b Annual Comparison Comparisons amplitude and with phase between of other scheme GPS 2 four from results ~ schemes c Table Comparison 3 of Lavallee et al.,2006 with SLR without ESA and SIO, including degree-2 deformation d Table 2 of Dong et al., 2003 from ~ , including degree-2 deformation Amplitude in X and Y direction agree with Lavallee, but the phase is different up Degree-two mass load has limited effects on the annual terms, only at the to 40 Amplitude degeree and phase of scheme 2 are larger in X direction; both amplitude level and The phase of phase sub-millimeter in of Y scheme direction 2 b and are phase clearly in smaller Z direction than show scheme good 2 in consistency X and Y direction, with encouraging SLR; Rotation but amplitude to parameter guess in that Z has direction phase littlein effects show X and large on Y the direction differences annual is terms likely not stable and that in earlier stage may be smaller e Average values of SLR results [Bouille et al., 2000; Chen et al.,1999; Cretaux et al., 2002; Moore and Wang, 2003] 18

19 Discussions TRF Realization with GNSS, data processing technique development for Multi-GNSS era, need steps forward TRF alignment to international standard, need push through more application projects for GNSS performance refining, IGS as an successful experience, we need make it forward. We realized the estimation of geocenter motion, annual amplitude and phase of by degree-one approach is consistent with those published Geocenter motion is the basic problem for TRF realization for high precision applications Multi-space technology is important to realize TRF for GNSS 19

20 THANK YOU!!! 20

(baking powder) 1 ( ) ( ) 1 10g g (two level design, D-optimal) 32 1/2 fraction Two Level Fractional Factorial Design D-Optimal D

$(baking powder) 1 ( ) ( ) 1 10g g (two level design, D-optimal) 32 1/2 fraction Two Level Fractional Factorial Design D-Optimal D$ ( ) 4 1 1 1 145 1 110 1 (baking powder) 1 ( ) ( ) 1 10g 1 1 2.5g 1 1 1 1 60 10 (two level design, D-optimal) 32 1/2 fraction Two Level Fractional Factorial Design D-Optimal Design 1. 60 120 2. 3. 40 10