Airborne Holographic SAR Tomography at L- and P-band O. Ponce, A. - PowerPoint PPT Presentation

Airborne Holographic SAR Tomography at L- and P-band O. Ponce, A. Reigber and A. Moreira. Microwaves and Radar Institute (HR), German Aerospace Center (DLR). 1

Outline • Introduction to 3-D SAR • Holographic SAR Tomography (HoloSAR) • Theory and Imaging Approaches • Experimental Realizations • Conclusions 2

Introduction – 3-D SAR Imaging SAR Interferometry (InSAR) SAR Tomography (SARTom) z 1 2 B . . 1 . z N 2 n n r r h h x x y y 3

Introduction – 3-D SAR Imaging SAR Interferometry (InSAR) SAR Tomography (SARTom) Retrieved Information: Retrieved Information: • Complex reflectivity • Height • Resolution in � • Single aspect angle • Single aspect angle Digital Elevation Model (DEM) of Iceland, 2011. 4

Introduction – 3-D SAR Imaging Circular SAR (CSAR) Holographic SAR Tomography (HoloSAR) z z 1 2 . . . n N n r r h x h y x y 5

Introduction – 3-D SAR Imaging Circular SAR (CSAR) Holographic SAR Tomography (HoloSAR) Retrieved Information: Retrieved Information: • Complex reflectivity • Complex reflectivity • Resolution in � • Resolution in � • Multiple aspect angles • Multiple aspect angles • Sub- � resolution in ��, �� • Sub- � resolution in ��, �� • Low resolution in � • High resolution in � Impulse Response Function - Luneburg Lens Impulse Response Function - Luneburg Lens 6

Introduction – Linear SAR VS Circular SAR Circular SAR Stripmap SAR 7

Experimental Realizations - Circular SAR – L-band Circular SAR Stripmap SAR Pauli basis, Coherent imaging, 500 m x 500 m, 0.06 m by 0.06 m sampling E ‐ SAR L ‐ Band, bandwidth 95MHz 8

Theory on HoloSAR – Impulse Response Function • Ambiguities • Resolution Gatelli, et al, The wavenumber shift in SAR interferometry, IEEE TGRS, 1994. Reigber, et al, First demonstration of Airborne SAR Tomography using Multi-baseline L-band data, IEEE TGRS , 2000. 9

Theory on HoloSAR – Impulse Response Function • ��, �� IRF • � IRF • Bandwidth enhancement F. Gatelli, et al, The wavenumber shift in SAR interferometry, IEEE TGRS, 1994. 10

Spectrum of HoloSAR - k space 11

Theory on HoloSAR – Impulse Response Function 19 tracks, ∆� � 12 m 1 track 3 tracks, ∆� � 150 m 12

Theory on HoloSAR – Imaging Approaches 1 2 3 O. Ponce, et al, Fully-Polarimetric High-Resolution 3-D imaging with CSAR at L-band, TGRS, 2014, in press . O. Ponce, et al, Analysis and optimisation of multi-circular SAR for fully polarimetric holographic tomography over forested areas, IGARSS 2013. O. Ponce, et al, Polarimetric 3-D Reconstruction from Multi-Circular SAR at P-band, GRSL 2014 . 13

Theory on HoloSAR – Imaging Approaches . . . . . . Compressive Sensing ( CS ) Beamforming ( BF ) . . . 1 2 3 Generalized Likelihood Ratio ( GLRT ) Coherent Addition - Fourier Incoherent Addition Holographic SAR Tomogram O. Ponce, et al, Fully-Polarimetric High-Resolution 3-D imaging with CSAR at L-band, TGRS, 2014, in press . O. Ponce, et al, Analysis and optimisation of multi-circular SAR for fully polarimetric holographic tomography over forested areas, IGARSS 2013. O. Ponce, et al, Polarimetric 3-D Reconstruction from Multi-Circular SAR at P-band, GRSL 2014 . 14

Experimental Realizations – HoloSAR at P-band F-SAR System 3-D HoloSAR Tracks Campaign Polarisations HH, HV, VH, VV Central Frequency P ‐ Band 20 MHz Chirp Bandwidth 500 Hz PRF 7 Circular passes 110 m Max. Baseline [m] 3800 m Radius avg. Region Vordemwald, CH. 15

Experimental Realizations – HoloSAR at P-band Pauli basis, �. � km diameter, �. �� m by �. �� m sampling 16

Forested area - � tracks – Span – ��, �� slices Subaperture, Fourier + Incoherent Fourier + Incoherent 17

Forested area - � tracks – Span – ��, �� slices CS + Incoherent Fourier + Incoherent 18

Forested area - � tracks – Span – ��, �� slices CS + Incoherent Fourier + Incoherent 19

Forested area - � tracks – Fourier + Incoherent – ��, �� slices Lexicographic (red line LIDAR) Span 20

Forested area - � tracks – CS + Incoherent – ��, �� slices Lexicographic (red line LIDAR) Span 21

Forested area - � tracks – CS + Incoherent – 3-D View 22

Experimental Realizations – HoloSAR at L-band F-SAR System 3-D HoloSAR Tracks Campaign Polarisations HH, HV, VH, VV Central Frequency L ‐ Band 50 MHz Chirp Bandwidth 500 Hz PRF 19 Circular passes 285 m Max. Baseline [m] 3700 m Radius avg. Region Kaufbeuren, DE. 23

Experimental Realizations – HoloSAR at L-band Pauli basis, �. � km diameter, �. �� m by �. �� m sampling 1) 15 x 15 x 50 m, 2) 300 x 300 x 50 m 24

Experimental Realizations – HoloSAR at L-band Single Tree – Pauli basis – 15 x 15 x 50 m 1 track 19 tracks, ∆� � 12 m 3 tracks, ∆� � 150 m 25 21

Experimental Realizations – HoloSAR at L-band Forested Area – �� Tracks - Pauli basis Compressive Sensing + Incoherent Compressive Sensing + GLRT Coherent 26 21

Forested area - �� tracks – Pauli - 2-D slices – � � ���. � m Compressive Sensing + GLRT Compressive Sensing + Incoherent Coherent * red line  LIDAR 27

Forested area - �� tracks – Pauli - 2-D slices – � � ���. � m Compressive Sensing + GLRT Compressive Sensing + Incoherent Coherent 28

Forested area – Pauli - �� tracks – 2-D slices - � � ���/��� m Compressive Sensing + GLRT Compressive Sensing + Incoherent Coherent 29

27 30 Compressive Sensing + Incoherent Forested area – Pauli - �� tracks – 3-D view

Conclusions • HoloSAR offers unique means to get the full 3-D backscattering over 360°. • Improvement of the effective BW by taking into account the several circular passes with vertical or horizontal separation • Theory is validated with Airborne acquisitions at L- and P-band over forests. • HoloSAR can be used as a powerful tool to measure biophisical parameters, and to reduce uncertainties of conventional 3-D SAR modes • Potential for Future Earth Observation Space Missions 31

L ‐ Band image with 6 cm sampling What are you seeing here? 32

DLR’s airborne SAR – L ‐ Band quad pol 1 2 3 4 1 3 2 4 Pauli basis, 1.8 km diameter, 0.06 m by 0.06 m sampling 33

Thanks for your attention! 34