differential code bias estimation for new signals and constellations 2014 - py08 -...
TRANSCRIPT
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 1
Differential Code Bias Estimation for New Signals and Constellations
O. Montenbruck, P. Steigenberger, A. Hauschild (DLR/GSOC)
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 2
Motivation
Differential Code Biases (DCBs) are required forCode-based positioningAmbiguity resolution using code observations
Current DCB products (CODE) limited to GPS/GLONASS legacy signalsNew signal biases introduced by additional signals and tracking modesBroadcast TGD/BGD and ISC parameters do not include all possible tracking modes
IGS needs to provide new multi-GNSS/multi-signal DCB productMulti-GNSS/multi-signal DCBs not presently supported in ionosphere+DCBprocessingAlternative, simplified method uses Global Ionosphere Maps for DCB estimation
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 3
DCB Determination using Global Ionosphere Maps
DCB from ionosphere-corrected pseudorange differenceUse of „known“ ionosphere based on global ionosphere maps
IGS IONEX productSingle-layer modelLimited accuracy
rcvsatSS2
S2S
SSSSSSSSSS
21
21
2121212121
DCB11STEC3.40
)()()()(
ff
εεMMBBIIPP
© ROB
© JPL
Measured Modelled Estimated
Averaged/ignored
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 4
Network and Receivers
MGEX Network (~80 stations) tracking GPS, GLO, GAL, (QZSS), BDSChallenge with modernized signals (GPS L2C+L5, Galileo E1+E5a):
Different receivers use different tracking modesExample: GPS L5 pilot-only tracking or combined pilot+data tracking
Period Jan. 1, 2013 – Mar 31, 2014Sampling period 1 minute
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 5
Processing Scheme
Select signal pair (S1,S2)
Determine satellite+receiver DCBsat+rcv from daily mean of ionosphere-corrected code difference
Determine individual DCBs for each satellite and station from combined DCBsfor all sites and satellites
assuming DCBsat+rcv = DCBsat+DCBrcv
applying a zero-constellation-mean constraint
Notes:Set of contributing stations depends on selected signalsPrototype implementation (IONDCB) enables processing of „all“ constellationsand dual-signal combinations but no combined adjustement of >2 signals(such as C1C, C1W, C2W)
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 6
DCB-Results: GPS L1-L2 P(Y) legacy DCB
L1-L2 P(Y) (C1W-C2W) DCBs compared to monthly CODE product for 01/2013Standard deviation of differences over entire constellation: 0.20 ns
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60G
023(
G32
)G
026(
G26
)G
033(
G03
)G
034(
G04
)G
035(
G30
)G
036(
G06
)G
038(
G08
)G
039(
G09
)G
040(
G10
)G
041(
G14
)G
043(
G13
)G
044(
G28
)G
045(
G21
)G
046(
G11
)G
047(
G22
)G
048(
G07
)G
049(
G27
)G
050(
G05
)G
051(
G20
)G
052(
G31
)G
053(
G17
)G
054(
G18
)G
055(
G15
)G
056(
G16
)G
057(
G29
)G
058(
G12
)G
059(
G19
)G
060(
G23
)G
061(
G02
)G
062(
G25
)G
063(
G01
)G
065(
G24
)
C1W
-C2W
Bia
s re
lativ
e to
CO
DE
[ns]
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 7
DCB-Results: Pilot vs. Pilot+Data Tracking GPS L5
GPS C1C-C5X
GPS C1C-C5Q
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 8
DCB-Results: Galileo
Galileo E1-E5a DCBs (squares: pilot tracking, diamonds: data+pilot tracking)DCB variation observed for newly launched satellites E19 and E20
-10.00
-8.00
-6.00
-4.00
-2.00
0.00
2.00
4.00
6.00
8.00
10.00
Jan 01 Jan 31 Mrz 02 Apr 01 Mai 01 Mai 31 Jun 30
Gal
ileo
DC
B(E
1-E
5a) [
ns]
E101(E11) E102(E12) E103(E19) E104(E20)E101(E11) E102(E12) E103(E19) E104(E20)
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 9
DCB-Results: BeiDou
DCBs for combinations of signals on bands B1 (C2I), B2 (C7I) and B3 (C6I)Monthly DCB solution and broadcast values for January 2013Consistency limited to 2ns
-20.00-18.00-16.00-14.00-12.00-10.00-8.00-6.00-4.00-2.000.002.004.006.008.00
10.0012.0014.0016.0018.0020.00
C00
1(C
30)
C00
3(C
01)
C00
4(C
03)
C00
5(C
04)
C00
6(C
06)
C00
7(C
07)
C00
8(C
08)
C00
9(C
09)
C01
0(C
10)
C01
1(C
05)
C01
2(C
11)
C01
3(C
12)
C01
4(C
13)
C01
5(C
14)
C01
6(C
02)
DC
B [n
s]
C2I-C7IC2I-C6IC7I-C6ITGD1-9nsTGD2+1ns
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 10
Site DCBs – Example Galileo
GAL C1C-C8QSeptentrio
Leica (old)
Leica (new) 70 ns
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 11
IGS MGEX DCB Product
Available at ftp://cddis.gsfc.nasa.gov/pub/gps/products/mgex/dcbSupported constellations and signals
GPS (C1C, C1W, C2W, C2X/L/S, C5Q/X)GLO (C1C, C1P, C2C, C2P)GAL (C1C/X, C5Q/X, C7Q/X, C8Q/X)BDS (C2I, C7I, C6I)
Prototype Bias SINEX formatTwo products
MGEXyyyy.bsx Satellite-only biases (weekly averages)MGEXyyyy_all.bsx Satellite and station biases (daily values)
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 12
Bias SINEX DCB Format Example%=BIA 0.01 IGS 14:110:29859 IGS 14:001:00000 00:000:00000 P 00000 0*-------------------------------------------------------------------------------------------------------* Solution INdependent EXchange Format (SINEX)*-------------------------------------------------------------------------------------------------------*-------------------------------------------------------------------------------------------------------+BIAS/SOLUTION*BIAS SVN_ PRN SITE DOMES____ OBS1 OBS2 BIAS_START__ BIAS_END____ UNIT __ESTIMATED_VALUE____ _STD_DEV___... DCB E101 E11 C1C C5Q 14:001:00000 14:005:00000 ns 4.2518 0.0496DCB E101 E11 C1C C5Q 14:005:00000 14:012:00000 ns 4.2838 0.0845DCB E101 E11 C1C C5Q 14:012:00000 14:019:00000 ns 4.2881 0.0725DCB E101 E11 C1C C5Q 14:019:00000 14:026:00000 ns 4.2985 0.1644DCB E101 E11 C1C C5Q 14:026:00000 14:033:00000 ns 4.3621 0.1399DCB E101 E11 C1C C5Q 14:033:00000 14:040:00000 ns 4.2716 0.1454DCB E101 E11 C1C C5Q 14:040:00000 14:047:00000 ns 3.9937 0.0000DCB E101 E11 C1C C5Q 14:047:00000 14:054:00000 ns 4.4921 0.1809DCB E101 E11 C1C C5Q 14:054:00000 14:061:00000 ns 4.0552 0.5291DCB E101 E11 C1C C5Q 14:061:00000 14:068:00000 ns 4.2821 0.2117DCB E101 E11 C1C C5Q 14:068:00000 14:075:00000 ns 4.1911 0.1618DCB E101 E11 C1C C5Q 14:075:00000 14:082:00000 ns 4.1363 0.1770DCB E101 E11 C1C C5Q 14:082:00000 14:089:00000 ns 4.1988 0.1490DCB E101 E11 C1C C5Q 14:089:00000 00:000:00000 ns 4.2597 0.0780DCB E101 E11 C1C C7Q 14:001:00000 14:005:00000 ns 4.5851 0.0738DCB E101 E11 C1C C7Q 14:005:00000 14:012:00000 ns 4.6154 0.1071DCB E101 E11 C1C C7Q 14:012:00000 14:019:00000 ns 4.5895 0.0769DCB E101 E11 C1C C7Q 14:019:00000 14:026:00000 ns 4.6393 0.1322DCB E101 E11 C1C C7Q 14:026:00000 14:033:00000 ns 4.6416 0.1322...-BIAS/SOLUTION*-------------------------------------------------------------------------------------------------------%=ENDBIA
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 13
DCB Product – Problems
Improper compensation of ionospheric delays using GIMsIndependence of DCBs for pilot-only vs. pilot+data tracking (no commonreceivers)Lack of observations for analysis of pilot-minus-data DCBsLacking self-consistency, i.e. DCB(a-b)+DCB(b-c) DCB(a-c)
Receiver dependence of satellite biasesImpact of multipath
BIAS SINEX is no standard (just prototype v0.01)
DLR/GSOC > IGS Workshop > 23-27 June 2014Slide 14
Summary and Conclusions
Simplified DCB estimation process based on differenced code observationsIonospheric delay eliminated using slant TEC from global ionospheric-map
Simpler algorithm than combined ionosphere and bias estimationPotentially less accurate
Allows computation of DCBs for modernized signals and new constellationsGood agreement of legacy L1-L2-P(Y) bias with CODE DCBs (σ=0.2ns)Different biases are estimated for different tracking modes
Potentially station-dependent effectsDependence on tracking mode / signal type still to be assessed
Comparison of DCBs with corresponding parameters of broadcast dataGood agreement for GPS and Galileo (<0.2 ns)Larger differences (0-12 ns) for BeiDou (further assessment required)