german aerospace center (dlr)microsoft powerpoint - pres_127_zink.ppt created date 12/10/2007...
TRANSCRIPT
Microwaves and Radar Institute – Hauke Fiedler – 29.11.2007
The TanDEM-X Mission: Overview and Status
H. Fiedler, M. Zink, G. Krieger, A. Moreira
German Aerospace Center (DLR)Microwaves and Radar InstituteOberpfaffenhofen
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
2015201420132012201120102009200820072006200520042003
TanDEM-X Timeline
German Call for Proposals for a Future Earth Observation Mission
Phase A Study
Selection of TanDEM-X for Phase A Study
Final Decision
TerraSAR-X Operation
TanDEM-X Operation
Phase B/C/D
At least 3 years of joint operation
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
NGA (NIMA) Standards for Digital Elevation Models Spatial Resolution Absolute Vertical Accuracy
(90%) Relative Vertical Accuracy (point-to-point in 1° cell, 90%)
DTED-1 90 m x 90 m < 30 m < 20 m
DTED-2 30 m x 30 m < 18 m < 12 m
HRTI-3 12 m x 12 m < 10 m < 2 m
HRTI-4 6 m x 6 m < 5 m < 0.8 m
absolute height error single point errors
(90% confidence interval)
point-to-point errors(90% confidence interval)
hΔ
hΔ⋅2~
hΔ
relative height errorDefinition of 90% point-to-pointerrors: 1°
1°
hHRTIh σ⋅≈Δ 33.2%90
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Available DEM Data Products
0
1
2
3
4
5
0 50 Coverage in Mio km²
Airborn
eLID
AR
Airborn
e SARPhotogram
metry
SRTM-C
(restr
icted
)
HR Satellit
es
Aster
SRTM-C
(free
)
SRTM-X
TanDEM-X
ERS 1/2
SPOT 5 HRS
DTE
D/H
RTI
Lev
elUSGS
GTOPO 30
global HRTI-3 DEM is a unique data product at a competitive price
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Comparison TanDEM-X vs SRTM
TerraSAR-XStripmap DEM
SRTMDEM
Reso.: 8m x 8mBeff: 270 mhambig: 17.2 mAcc.: ~0.8 m
Reso.: 90m x 90mBeff: ~38 mhambig: ~175 mAcc.: ~16 m
Pictures Courtesy from IGARSS07: TerraSAR-X Interferometry: Report on a First Assessment (N. Adam et al.)Obs. Site: R. Metzig (HR)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
both satellites transmit and receive independentlytemporal decorrelation & atmospheric disturbances
backup solution
Pursuit Monostatic
transmitter alternatesbetween PRF pulses
provides two baselinesenables synchronisation, calibration & verification
one satellite transmits and both satellites receive simultaneouslydual use of signal energy
requires synchronisation
Alternating BistaticBistatic
TanDEM-X Data Acquisition Modes
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Impact of Oscillator Noise
increasedsidelobes
mainlobedispersion
azimuthdisplacement
interferometric phase errors
m
m
X
Txτ
Rxτ
independentoscillators
Tx
Rx
≠
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Phase Referencing in TanDEM-X
phase referencing can achieveshort term rmse below 1°
Synchronisation LinkSynchronisation Link Analysis of Residual ErrorsAnalysis of Residual Errors
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Tx Waveform Diversity for Synchronisation
Rx1f
Tx1
Rx2f
Tx2
*
gλ
2λ1λ
Rx1
Rx2
*
second bistatic second bistatic interferograminterferogram
*
firstfirst bistatic bistatic interferograminterferogram
*
monostatic monostatic interferograminterferogram
Rx1f
Tx1
Rx2f
Tx2
*
gλ
2λ1λ
Rx1
Rx2
*
second bistatic second bistatic interferograminterferogram
*
firstfirst bistatic bistatic interferograminterferogram
*
monostatic monostatic interferograminterferogram
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Collision Avoidance
HELIX satellite formation enables safe operationhorizontal cross-track separation at equator by different ascending nodesvertical (radial) separation at poles by orbits with different eccentricity vectors
horizontalbaseline
verticalbaseline
NH(desc.)
SH(asc.)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
linear phaseAntenna Tapering
12 m x 12 mPost Spacing
28-32 kmSwath Width
< 1 kmAlong-Track Disp..
500 mBaseline (perp.)
Ulaby (X-Band, VV, Soil)
Sigma Nought Model (90% occurence)
3 bit (BAQ)Quantization
1/10 pixelMis-Registration
2266 HzProcessed Bandwidth
~3500 HzPRF (swath variant)
4.8 m x 0.7 mAntenna Size (Tx, Rx)
4.1 dBLosses (rad., atm.,..)
4.3 dBNoise Figure TRM
18 %Duty Cycle
2260 WPeak Transmit Power
100 MHzChirp Bandwidth
0.031 mWavelength
Relative Height AccuracyValueParameter
DEM Performance Prediction (Bistatic Stripmap Mode)
90% 90% pointpoint--toto--pointpoint errorserrors
σ (standarddeviation)
BB⊥⊥ = 500 m= 500 m
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
linear phaseAntenna Tapering
12 m x 12 mPost Spacing
28-32 kmSwath Width
< 1 kmAlong-Track Disp..
1000 mBaseline (perp.)
Ulaby (X-Band, VV, Soil)
Sigma Nought Model (90% occurence)
3 bit (BAQ)Quantization
1/10 pixelMis-Registration
2266 HzProcessed Bandwidth
~3500 HzPRF (swath variant)
4.8 m x 0.7 mAntenna Size (Tx, Rx)
4.1 dBLosses (rad., atm.,..)
4.3 dBNoise Figure TRM
18 %Duty Cycle
2260 WPeak Transmit Power
100 MHzChirp Bandwidth
0.031 mWavelength
Relative Height AccuracyValueParameter
DEM Performance Prediction (Bistatic Stripmap Mode)
BB⊥⊥ = 1000 m= 1000 m
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Height of AmbiguityTanDEM-X enables large baselines which allow for ultra high resolution DEMs with height accuracies in the sub-meter range, but …
πϕΔ 2=(height of ambiguity)
Compromise on Accuracy for Global DEMuse reduced baselinesadditional acquisitionsfor difficult terrain
acquisition scenariofor global DEM according to HRTI-3
Local/Regional Ultra High Resolution DEMs
multiple data acquisi-tions with large and small baselines
regional DEMs withsub-meter resolution(e.g. HRTI-4)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Data Acquisition Plan for Global DEM
•• map total Earth landmass with map total Earth landmass with two different baselinestwo different baselines
•• additional acquisitions in difficult additional acquisitions in difficult and mountainous terrain (~20 %)and mountainous terrain (~20 %)
BB11 BB22
m9 m45 m 30
011
1
213
2
1 ≈−
=⎭⎬⎫
≈≈
hhh
hh
0°
10°
20°
30°
40°
50°
60°
70°
1 32 54 76 8 9Beam Number
Latit
ude
HelixHelix aaΔΔee aaΔΔii ΨΨ
HH11 386 m386 m 330 m330 m 200200°°HH22 440 m440 m 385 m385 m 210210°°
HH33 440 m440 m 550 m550 m 210210°°
Optimized Set of HELIX Formations
Multiple DEM Acquisitions
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Data Acquisition Scenario: Example
First Mission PhaseSecond Mission
PhaseThird Mission Phase
Different Mission Phases with its Helices (i.e. optimised height of ambiguity) will provide smooth performance over the Earth’s surface at all latitudes
Consequence: Final DEM will be available at end of mission
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Combination of Interferometric DEM Acquisitions
hamb = 45 m
hamb = 30 m
optimum combination of all swaths
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
DEM Performance (flat terrain: rock & soil)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
DEM Performance (20 % slopes)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
DEM Performance (20 m vegetation)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Baseline Estimation and DEM Calibration• Both satellites are exposed to almost
identical orbit perturbations
- negligible azimuth modulation / twisting of DEM swath (ΔB < 0.25 mm for 500 km swath and ‘unmodelled’ δa < 100 nm/s2)
- vertical bias and tilt of raw DEM swaths due to initial baseline estimation errors
• Precise baseline estimation by
- double-difference GPS carrier-phase measurements
- accurate orbit propagation model
- several studies predict a 3-D accuracy in the order of 1-2 mm
baselinevector
jAρ
jBρ
kAρ
kBρ
j k
GPS satellites
DEMcalibration with:- swath overlaps- crossing orbits- ICESat- GPS Tracks- ...
BBestimatedestimated
BBtruetrue
""biasbias""
ΔΔBB
BBestimatedestimated
BBtruetrue
""biasbias""
ΔΔBB
similar orbit perturbations:
- drag, - Earth,- sun,- moon,- ...
relative orbit model
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Secondary Mission Goals & New Techniques
S I R E V R a w D a ta
C h i r p S c a li n g
R a n g e F F T
R a n g e IF F T
R C M C o r r e c t i o n
R a n g eC o m p r e s s i o n
S e c o n d a r y R a n g eC o m p r e s s i o n
A z i m u t h F F TP h a s e C o r r e c t i o n
A z i m u t h S i d e -l o b e S u p r e s s i o n
A z i m u t h I F F T
A zi m u t h F F T
S I R E V I m a g e
L O S - C o r r e c t i o n a n d F i r s t O r d e r M o t i o n C o m p e n s a t i o n
A z i m u t h F F T
A zi m u t h I F F T
S e c o n d O r d e r M o t i o nC o m p e n s a t i o n
B0
CA
T s
τ
t
0f a
AC
Bτ
f a0
f r
C B A
I n t e g e rR e s a m p li n g
S I R E V R a w D a ta
C h i r p S c a li n g
R a n g e F F T
R a n g e IF F T
R C M C o r r e c t i o n
R a n g eC o m p r e s s i o n
S e c o n d a r y R a n g eC o m p r e s s i o n
A z i m u t h F F TP h a s e C o r r e c t i o n
A z i m u t h S i d e -l o b e S u p r e s s i o n
A z i m u t h I F F T
A zi m u t h F F T
S I R E V I m a g e
L O S - C o r r e c t i o n a n d F i r s t O r d e r M o t i o n C o m p e n s a t i o n
A z i m u t h F F T
A zi m u t h I F F T
S e c o n d O r d e r M o t i o nC o m p e n s a t i o n
B0
CA
T s
τ
t
0f a
AC
Bτ
f a0
f r
C B A
I n t e g e rR e s a m p li n g
S I R E V R a w D a ta
C h i r p S c a li n g
R a n g e F F T
R a n g e IF F T
R C M C o r r e c t i o n
R a n g eC o m p r e s s i o n
S e c o n d a r y R a n g eC o m p r e s s i o n
A z i m u t h F F TP h a s e C o r r e c t i o n
A z i m u t h S i d e -l o b e S u p r e s s i o n
A z i m u t h I F F T
A zi m u t h F F T
S I R E V I m a g e
L O S - C o r r e c t i o n a n d F i r s t O r d e r M o t i o n C o m p e n s a t i o n
A z i m u t h F F T
A zi m u t h I F F T
S e c o n d O r d e r M o t i o nC o m p e n s a t i o n
B0
CA
T s
τ
t
0f a
AC
Bτ
f a0
f r
C B A
I n t e g e rR e s a m p li n g
Pol-InSAR
Digital Beamforming
Bistatic Observations
Multi BaselineInSAR
Super Resolution
Along-Track Interferometry(fully polarimetric !)
(4 phase centres !)
(large bandwidth !) (bistatic angle !)
(HELIX formation !)
(flexible baselines !)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Along-Track Interferometry
Amplitude:Amplitude:
Interferometric Interferometric Phase:Phase:
r+Δr r
t+Δt t
Balong
accurate measurements of radial displacement betweentwo radar observations separated by a short time lag
(Ameland, Holland)
HELIX
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
ATI Performance: Measurement of Ocean Currents
vwind = 15 m/sτe = 4 ms
NRCS = -12.1 dB
vwind = 10 m/sτe = 6 ms
NRCS = -15.6 dB
vwind = 5 m/sτe = 12 ms
NRCS = -21.5 dBΔΔv v < 0.1 < 0.1 m/sm/s @ @ 50 m x 50 m50 m x 50 m
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Along-Track Interferometry
Interferometric Interferometric Phase:Phase:
HELIX
sensitive to fast movements
sensitive to slow movements
split antenna
0.2 km/h~9 km/hΔv
12 km/h500 km/hvamb
45°θinc
- 12 dBm2/m2σ0
12 m12 mΔx
100 m2.4 mBalong
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Digital Beamforming with Four Phase Centres
without reconstruction with reconstruction
Ch. 2Ch. 1
SAR Proc.
P2(f) P3(f)P1(f) P4(f)
EnablesHigh
ResolutionWide Swath
Imaging
Ch. 3 Ch. 4
AmbiguitySuppression
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
“Double Differential SAR Interferometry”
Grounding line detection, vegetation growth, snow/ice accumulation, … ?
h(t1)
Δh ~ ϕ2 - ϕ1
coherence between passes not mandatory
ϕ1pass 1
pass 2
Δh < 10 cm h(t2)
ϕ2
Bistatic Strip mapB = 3000 m Δx = 12 m
Bistatic Strip mapB = 3000 m Δx = 12 m
e.g. difference between two single-pass cross-track interferograms
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
Polarimetric SAR Interferometry
polarisations
pdf(μmin)
pdf(μmax)
Parametersh = 1.2 mβ = 4.0 dB/mμmin = -7.0 dBμmax = 3.0 dBB⊥ = 4 kmθinc = 35°Δx = 30 m
Parametersh = 1.2 mβ = 4.0 dB/mμmin = -7.0 dBμmax = 3.0 dBB⊥ = 4 kmθinc = 35°Δx = 30 m
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
RxθTxθ
φ
Extended Observation Space
Bistatic Radar Imaging
Color Color compositecomposite of of threethree bistatic SAR bistatic SAR imagesimagesfromfrom DLRDLR--ONERA bistatic SAR ONERA bistatic SAR experimentexperiment
(P. (P. DuboisDubois Fernandez et al. , IEE Fernandez et al. , IEE ProcProc. RSN, 2006). RSN, 2006)improved detection, segmentation, and classification from bistatic RCSnew image and object parameters (bistatic Doppler, multiple shadows, ...)
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007
TanDEM-X Summary
TanDEM-X is the next German Earth Observation Mission to be launched in 2009
TanDEM-X has outstanding scientific and commercial potentials
TanDEM-X key technologies are:
bistatic radar operation and phase synchronisation
precise baseline determination
close formation flying capability
new algorithms for interferometric processing
TanDEM-X plays a key role in the development of next generation bistatic and multistatic SAR missions and applications
FRINGE 2007Microwaves and Radar Institute- Hauke Fiedler – 29.11.2007