First TOPSAR image and interferometry results with TerraSAR-X

Transcription

1 First TOPSAR image and interferometry results with TerraSAR-X A. Meta, P. Prats, U. Steinbrecher, R. Scheiber, J. Mittermayer DLR Folie 1 A. Meta

2 Introduction Outline TOPSAR acquisition mode First TerraSAR-X TOPSAR images First TerraSAR-X TOPSAR interferometric fringe results Conclusions and future work Folie 2

3 Introduction TOPSAR is a new acquisition mode proposed by E. Attema (ESA- ESTEC) and F. Rocca (POLIMI) Developed by POLIMI 1 TOPSAR requires a fast rotation of the azimuth antenna pattern it aims at achieving the same coverage and resolution as ScanSAR, but with a nearly uniform SNR and DTAR It is going to be the operational mode for Sentinel-1 Interferometric Wide Swath (IWS) mode It is being successfully demonstrated with TerraSAR-X for the first time within the framework of an ESA project 1) F. D. Zan and A. M. Guarnieri, TOPSAR: terrain observation by progressive scan, IEEE Trans. Geosci. Remote Sensing, vol. 44, no. 9, pp , Sept Folie 3

4 TOPSAR acquisition mode Folie 4

5 TOPSAR acquisition mode Folie 5

6 Antenna array effects Boresight case Normalized angle of the steering angle equal to 0.2 Folie 6

7 Antenna array effects The azimuth array antenna pattern is amplitude weighted by the single element pattern When the array pattern is being steered, the effect of the grating lobes is increasing. A small residual scalloping effect can be introduced However, it is much smaller than the equivalent ScanSAR case and has a lower frequency due to extended burst image area Folie 7

8 TerraSAR-X TOPSAR analysis A DTAR variation and scalloping of less than 1 db is expected in a typical TerraSAR-X TOPSAR configuration Folie 8

9 Inverse TOPSAR Nominal TOPSAR Inverse TOPSAR k R k R 2v 1 1! 0 i! 0 + = " # ki! = + k! v v R0 The timeline equation are the same for TOPSAR and inverse TOPSAR Steering rate higher in inverse TOPSAR, therefore higher scalloping and DTAR variation to be expected Folie 9

10 First TOPSAR images with TerraSAR-X Folie 10

11 TerraSAR-X TOPSAR configuration TerraSAR-X parameters four subswaths ~80 km swath coverage 16 m azimuth resolution Carrier frequency Along-track antenna length TR modules in along-track One-way azimuth beamwidth (3 db) Satellite velocity Height Maximum steering angle 9.65 GHz 4.8 m deg 7608 m/s 514 km ± 0.75 deg Folie 11

12 TerraSAR-X TOPSAR acquisition example SS 1 SS 2 SS 3 SS 4 Input parameters Mean look angle - [deg] Slant middle range [km] Ground velocity [m/s] PRF [Hz] Calculated parameters Integration beamwidth [deg] Steering angle rate [deg/s] DTAR [db] Burst Time [s] Target Doppler Bandwidth [Hz] Maximum steering angle - [deg] Number of Pulses per Burst [ ] Burst image length [m] Folie 12

13 Vendome: inverse TOPSAR image Scene size: 84 km in azimuth and 72 in slant range 8 bursts Antenna steering from deg to deg Burst image steering from deg to deg Small scalloping effect visible No scalloping correction applied Data Take acquired on June 29 th, 2007, only two weeks after the TerraSAR- X launch Folie 13

14 Neuville: TOPSAR image Scene size: 90 km in azimuth and 75 in slant range 9 bursts Antenna steering from deg to deg Burst image steering from deg to deg Difficult to see scalloping effects No scalloping correction applied Folie 14

15 Azimuth profiles Inverse TOPSAR azimuth profile example ~0.8 db scalloping Nominal TOPSAR azimuth profile example ~0.4 db scalloping Folie 15

16 Challans: Google image Folie 16

17 Challans: TOPSAR image Time acquisition: July 9 th, 2007 at 6.26 am Folie 17

18 Challans: ScanSAR image Time acquisition: September 2 nd, 2007 at 6.26 am Folie 18

19 First TOPSAR Interferometry results with TerraSAR-X Folie 19

20 A squint error!" TerraSAR-X TOPSAR interferometry The TerraSAR-X characteristics: translates into a Doppler shift Along-track position accuracy is within ~50 m (0.005 deg at 620 km) Maximum Doppler shift of approximately 40 Hz 2v f shift =!" # For a processed Doppler bandwidth of 455 Hz (16 m res.), a maximum interferometric resolution loss of less than 9% is expected due to band filtering. Accurate coregistration is required due to high Doppler centroids at the edges of the burst images. Folie 20

21 Uyuni salt lake Folie 21

22 Uyuni salt lake Folie 22

23 Uyuni salt lake: TOPSAR master image Scene size: 110 km in azimuth and 74 in slant range 8 bursts Look angle is varying from 32.9 degree to 37.9 degree Folie 23

24 Uyuni salt lake: TOPSAR slave image Same configuration as the master data take acquisition Coarse misalignment of 70 pixel in range and 3 in azimuth Along-track misalignment around 22 m. Folie 24

25 Uyuni salt lake: TOPSAR coherence image Folie 25

26 Uyuni salt lake: TOPSAR interferogram The estimated range frequency is linearly decreasing with increasing range and with along-track position, estimated perpendicular baseline varying from 67 to 38 meter The resulting ambiguity height is varying from approximately 92 to 173 meter. Folie 26

27 Uyuni salt lake: TOPSAR final interferogram Folie 27

28 Unwrapped phase Folie 28

29 Doppler centroid variation effects # 4"! $! f ( r ) 2 % & $ % ' & * 2v + ' * + DC 0 rg _ err = ( rmis 1) 1) & ' Negligible effect! = " # t az _ err 2 fdc Very important 0.1 pixel azimuth misregistration Folie 29

30 Next steps Calibration of the TOPSAR interferometry phase and quantitative analysis Processing of TerraSAR-X data takes for Sentinel-1 simulations in terms of coverage, steering angle, scalloping Final results will be used for suggestions for the TOPSAR IWS of the ESA Sentinel-1 sensor Folie 30

31 Conclusions Folie 31

32 Conclusions TerraSAR-X TOPSAR data takes have been successfully generate, commanded and executed First TOPSAR images have been produced with the experimental processor developed at DLR First TOPSAR interferometric fringe have been successfully generated over the Uyuni salt lake Unique validation approach by joining the knowledge and expertise of TerraSAR-X SEC, processing and airborne groups at the DLR Folie 32

33 Thanks for your attention Folie 33