quality disclaimersentinel -1 no. 02...sentinel-1a tile #5 intermittent failure: impact on l1...

1

Quality Disclaimer

Description:

Failure on Tile amplifier #5 of the receiving antenna from 18/10/2014 to 20/01/2015.

Degradation types:

DEGRADED_PRODUCT_RADIOMETRY DEGRADED_PRODUCT_GEOLOCATION

DEGRADED_RADIOMETRIC_CALIBRATION DEGRADED_PLATFORM_POINTING

DEGRADED_ORBIT_CONTROL DEGRADED_PERFORMANCE_INSTRUMENT_ANOMALY

COMPLETE_PRODUCT_DEGRADATION SLICE_PRODUCT_NON_CONCATENABLE

OTHER

Degradation percentage1:

0.0

Impacted products:

Platform: S-1A S-1B

Acquisition mode: EW IW SM WV RF

Product type: RAW SLC GRD OCN

Resolution class: MR HR FR N/A

Polarization: SH (Single pol. H) SV (Single pol. V)

DH (Double pol. H) DV (Double pol. V)

HH HV VV VH

Processing facility: PAC1 / UPA PAC2 / DPA

CGS1 / Matera CGS2 / Svalbard CGS3 / Maspalomas

IPF version: 002.36

Instrument Configuration ID (RDB): 3

ADF files:

AUX_INS N/A

AUX_CAL N/A

AUX_PP1 N/A

AUX_PP2 N/A

Beginning of the issue:

Start acquisition date: 2014-10-18 15:29:30 UTC

Start generation date: 2014-10-18 19:02:02 UTC

Orbit: 2886

Datatake (hex): 003445

End of the issue:

not yet defined available

End acquisition date: 2015-01-20 19:04:54 UTC

End generation date: 2015-01-20 19:23:04UTC

Orbit: 4259

Datatake (hex): 0052D3

Cause: Failure on Tile Amplifier #5: the consequence is all the TRMs sharing the same Tile Amplifier are failing in both Rx H and V polarisations.

Status:

Slight degradation of the radiometric characteristics products due to slight change of antenna patterns: worse azimuth ambiguity ratios due to higher azimuth sidelobes of 0.3 to 4.4 dB depending on the acquisition mode & sub-swath and a reduction in the NESZ due to loss of power (by approx 0.7 dB).

1 Percentage of degradation of the data in the product (100% means that the product should be masked in the product catalogue)

SENTINEL-1 No. 02

2

References:

MPC ref: MPC-692

PDGS ref: S1PDGS-14169

ARTS ref: GS1_SC-79

MPC-S1

Sentinel-1A Tile #5 intermittent failure: Impact on L1 product quality

Reference: MPC-204

Nomenclature: OI-MPC-OTH

Issue: 1. 0

Date: 2015,Jan.30


MPC-204 OI-MPC-OTH V1.0 2015,Jan.30 i.1

Proprietary information: no part of this document may be reproduced divulged or used in any form without prior permission from CLS. F

ORM

-NT-G

B-1

0-1

Chronology Issues:

Issue: Date: Reason for change: Author

1.0 30.01.2015 First Issue MPC-S1

People involved in this issue:

Written by (*):

MPC-S1-team Date + Initials:( visa or ref)

Checked by (*):

P.J Meadows Date + Initial:( visa ou ref)

P.J.Meadows

Approved by (*): G.Hajduch Date + Initial:( visa ou ref)

G.Hajduch

Application authorized by (*):

N. Miranda Date + Initial:( visa ou ref)

*In the opposite box: Last and First name of the person + company if different from CLS

Index Sheet:

Context: Investigation on L1 data quality related to antenna failures

Keywords: S-1A, Antenna, Tile, Failure

Hyperlink: [Link to the document on SharePoint]

https://shareesa.cls.fr/MPC-S1/Technical/Forms/AllItems.aspx?RootFolder=%2FMPC%2DS1%2FTechnical%2FTechnical%20%28All%29%2FOI%2DMPC%2DOTH%2D0204%20Sentinel%2D1A%20Antenna%20failures%20impacts&FolderCTID=0x0120006AC9C3F8E1775248AD23E33379BEBFC6&View=%7bB21F61AA-90C7-4A26-8FE8-674DD396E651%7d




ORM

-NT-G

B-1

0-1

List of tables and figures

List of tables:

Table 1: main events related to antenna and data quality since launch ................................ 1

Table 2: Azimuth processing band and Hamming coefficients considered for performance evaluation. .................................................................................................... 20

Table 3: Azimuth processing band and Hamming coefficients considered for performance evaluation. .................................................................................................... 20

List of figures:

Figure 2: H-polarization error matrixes computed the 15-10-2014 (Top) and the 10-01-2015 (Bottom) . ....................................................................................................... 2

Figure 3: Evolution over time of the antenna RX-coefficients averaged over columns (RFC processing). Top: H-pol, Bottom: V-pol .................................................................. 3

Figure 4: Evolution of the PG product (1/PG) during anomaly and intermittent recovery periods ........................................................................................................... 4

Figure 5: Long term PG monitoring .............................................................................. 4

Figure 6: Error Matrix from the RFC product in RxV one day-apart ...................................... 5

Figure 7 Examples of the comparisons between the real and nominal azimuth antenna patterns. The plots on the right are the same as those on the left, but they are zoomed in over the main beam. Blue lines are the nominal patterns, red are the real patterns and green is the difference between the nominal and real patterns. ............................. 6

Figure 8 The differences between the nominal and real azimuth antenna patterns for all swaths and polarisations for the SM, IW and EW modes. ............................................. 7

Figure 9 Comparison between the nominal and real elevation antenna patterns. The plots on the left shows the real and nominal patterns for the HH polarisation. The plots on the right show the difference between the real and nominal elevation antenna patterns for all polarisations. ............................................................................................... 8

Figure 10: Difference between azimuth patterns of IW TopSAR beams for different ABT indexes. ........................................................................................................ 10

Figure 11: Difference between azimuth patterns of EW beams for different ABT indexes. ...... 11

Figure 12: Doppler Calibration Profile for IW beams before (top figure) and after (bottom figure) tile 5 failure. ........................................................................................ 12

Figure 13: Comparison between currently used de-scalloping pattern (blue line) and de-scalloping pattern calculated with S1 CFI AM (red dashed line) for IW beams. ............... 13

Figure 14: Comparison between currently used descalloping pattern (blue line) and descalloping pattern calculated with S1 CFI AM (red dashed line) for TopSAR EW beams .. 15

Figure 15: AAP phase before (top) and after (bottom) tile 5 failure for Stripmap beams. ........ 16

Figure 16 Gamma profiles calculated from L1 products. Top left is S1, top right is S6, mid left is IW1, mid right is IW3, bottom left is EW1. The red and black lines indicate the different polarisations. ..................................................................................... 17

Figure 17: (a) Transponder D39 (Ochsenhausen) RGC image intensity. (b) Selection of 7 range samples and 1.2*PRF azimuth lines around the PT maximum. (c ) Estimated 2-w Azimuth Antenna Pattern from data. ............................................................................... 18

Figure 18: Comparison between data profile (blue) and AAP from the AM, Stripmap S6 case. .. 19




ORM

-NT-G

B-1

0-1

Figure 19: Comparison between data profile (blue) and AAP from the AM, IW1 case. ............. 19

Figure 20: Theoretical NESZ performance: comparison between expected values with no-failures (dashed line) and with (solid line) the antenna failures as in Figure 1. ............. 20

Figure 21: Effect of intermittent recovery happening during a data-take ............................ 21

Figure 22: Verification of the TOPS de-scalloping correction made for 2 EW product after (left) and before (right) the anomaly. .......................................................................... 23

Applicable documents

AD 1 None.

Reference documents

RD-1 Antenna Status: RFC and Internal Calibration Analyses, Aresys presentation, 21st October 2014 S1A_RFC_ARESYS_20141021.pdf

RD-2 Sentinel-1 MPC Quality Disclaimer No.2, 5th Nov 2014.

RD-3 Antenna Status: RFC and Internal Calibration Analyses, Aresys presentation, 20th May 2014 S1A_CP_RFC_ARESYS_20140520

RD-4 Sentinel-1 Antenna Model Analysis, BAE Systems ATC, 23rd January 2015, BAE_AntennaModelAnalysis_23Jan2015.pdf

RD-5 Sentinel-1 Product Analysis - Amazon Image Analysis, BAE Systems ATC, 27th January 2015, BAE_Product_Analysis27Jan2015_Amazon.pdf




ORM

-NT-G

B-1

0-1

List of Contents

1. Purpose and scope ........................................................................ 1

2. Introduction ................................................................................ 1

2.1. Sentinel-1A antenna status overview since launch (RFC time series, error matrices) ............................................................................. 1

2.1.1. Internal calibration and intermittent power recovery ...................................... 3

2.2. Antenna pattern shape comparison, ideal vs. with-failures: ............... 5

2.2.1. Azimuth Antenna Patterns (AAPs) ............................................................... 5

2.2.2. Elevation Antenna Patterns (EAPs) .............................................................. 7

2.2.3. Steered AAP and descalloping function ......................................................... 9

2.3. AAP phase ramp targets shifting effects ....................................... 15

3. Impact on L1 data products quality ................................................... 17

3.1. EAP from AM: consistency with Gamma profiles ............................. 17

3.2. AAP from AM: whitening verification ........................................... 18

3.3. Impact of antenna failures: update of the expected theoretical performance (NESZ and ambiguity) ................................................... 19

3.4. Impact of tile-5 intermittent power recovery on radiometric quality ... 21

3.5. TOPS De-scalloping verification .................................................. 22

4. Conclusions ................................................................................ 24

Appendix A - List of acronyms ............................................................ 25


MPC-204 OI-MPC-OTH V1.0 2015,Jan.30 1

Proprietary information: no part of this document may be reproduced divulged or used in any form without prior permission from CLS.

FO

RM

-NT-G

B-7

-1

1. Purpose and scope

The present technical note focuses on the Sentinel-1A antenna status and evolution since launch and analyzes the impacts of such evolution on the SAR data quality.

The document collects the relevant analyses and results generated in the framework of the Sentinel-1A Mission Performance Centre up to the 20th of January 2015, with the purpose of giving a clear picture of the L1 data quality.

2. Introduction

2.1. Sentinel-1A antenna status overview since launch (RFC time series, error matrices)

The Sentinel-1A antenna is routinely monitored through the processing and analysis of the RFC mode data. Since launch, the following events have been recorded.

Date Event reference

03-04-2014 S1-A Launch -

05-05-2014 Tile 4, rows 11 and 12 TRM failures (6 elements)

RD-3

09-06-2014 Tile 4, rows 11 and 12 TRM failures (1 element)

RD-3

18-10-2014 Tile 5, all TRM failures (Tile 5, RX only, both polarizations)

RD-1

05-11-2014 Issue of a Quality Disclaimer RD-2

Table 1: main events related to antenna and data quality since launch




FO

RM

-NT-G

B-7

-1

Figure 1 below reports the TX and RX error matrixes computed on the 15-10-2014 and on the 10-01-2015.

Figure 1: V-polarization error matrixes computed on the 15-10-2014 (Top) and on the 10-01-2015 (Bottom) .

Figure 2: H-polarization error matrixes computed the 15-10-2014 (Top) and the 10-01-2015 (Bottom) .




FO

RM

-NT-G

B-7

-1

The failure of tile 5 in RX can be also noticed by looking at the average RFC coefficients power, computed over each tile and plotted over time (see Figure 3 below).

Figure 3: Evolution over time of the antenna RX-coefficients averaged over columns (RFC processing). Top: H-pol, Bottom: V-pol

2.1.1. Internal calibration and intermittent power recovery

The effect of the tile#5 anomaly impacts the PG, which shows a power variation proportional to the number of failures, in this case a reduction close to 13/14 (0.7 dB) of the nominal value.

The 1/PG is used to normalise the L1 product. It has been seen, back in October 2014, that the PG is impacted by tile 5 failure. For IW, which is the main mode operation, a PG change of 0.8dB has

09-APR-2014 21-JUN-2014 02-SEP-2014 14-NOV-2014 26-JAN-2015

-40

-35

-30

-25

-20

-15

-10

-5

0

RX H abs mean [dB]


-150

-100

-50

0

50

100

150

RX H phase mean [deg]

Tile 1

Tile 2

Tile 3

Tile 4

Tile 5

Tile 6

Tile 7

Tile 8

Tile 9

Tile 10

Tile 11

Tile 12

Tile 13

Tile 14

Tile 1

Tile 2

Tile 3

Tile 4

Tile 5

Tile 6

Tile 7

Tile 8

Tile 9

Tile 10

Tile 11

Tile 12

Tile 13

Tile 14


-30

-25

-20

-15

-10

-5

0

RX V abs mean [dB]


-20

0

20

40

60

80

100

120

140

160

RX V phase mean [deg]

Tile 1

Tile 2

Tile 3

Tile 4

Tile 5

Tile 6

Tile 7

Tile 8

Tile 9

Tile 10

Tile 11

Tile 12

Tile 13

Tile 14

Tile 1

Tile 2

Tile 3

Tile 4

Tile 5

Tile 6

Tile 7

Tile 8

Tile 9

Tile 10

Tile 11

Tile 12

Tile 13

Tile 14




FO

RM

-NT-G

B-7

-1

been observed as shown in Figure 4. Figure 4 also shows that in some occasions the PG recovers its original level.

Figure 4: Evolution of the PG product (1/PG) during anomaly and intermittent recovery periods

In practice, an evolution of the PG captured by the internal calibration pulses is compensated by the IPF. This has been verified as there is no obvious degradation over the RCS measurements performed over the calibration sites (e.g. DLR transponders or rain-forest).

The PG product is monitored over the mission lifetime. It is possible to identify epochs of temporary recovery from the PG long term monitoring. Further events reported in 12/01/2015 and 20/01/2015 are not displayed in Figure 5.

Figure 5: Long term PG monitoring

Those periods of recovery are also captured by the RFC that is going back to nominal.




FO

RM

-NT-G

B-7

-1

(a) Error Matrix from 20150118 (b) Error Matrix from 20150119

Figure 6: Error Matrix from the RFC product in RxV one day-apart

For all these events of intermittent recovery, it is expected to see radiometric jumps from slice to slice for one or several consecutives slices (see section “Impact of tile-5 intermittent power recovery on radiometric quality” below).

2.2. Antenna pattern shape comparison, ideal vs. with-failures:

The Antenna Model has been used to generate patterns which represent the current state of the antenna (referred to as real), and these have been compared to the patterns for the perfect case (referred to as nominal).

The real patterns have been generated using:

A failure matrix, with includes failures for the current failed modules in tile 4, and the

failures for tile 5 (as described above)

An error matrix which represents the current state of the antenna

The nominal patterns have been generated using:

A failure matrix with no modules indicated as failed

An error matrix which indicates all modules are working perfectly

The following two sections describe the impact for the Azimuth Antenna Patterns and the Elevation Antenna Patterns respectively.

2.2.1. Azimuth Antenna Patterns (AAPs)

Figure 7 shows some examples of the comparison between the real and nominal patterns. The comparisons for the other modes and polarisations are consistent with those shown here. This consistency is shown in Figure 8 which shows the difference between the nominal and real patterns for SM, IW and EW modes over the main beam. These figures show there is a change in the shape of the azimuth antenna patterns over the main beam, and a significant difference in their side lobes. Azimuth antenna patterns for a couple of other swaths can be found in RD-4.




FO

RM

-NT-G

B-7

-1

Figure 7 Examples of the comparisons between the real and nominal azimuth antenna patterns. The plots on the right are the same as those on the left, but they are zoomed in

over the main beam. Blue lines are the nominal patterns, red are the real patterns and green is the difference between the nominal and real patterns.




FO

RM

-NT-G

B-7

-1

Figure 8 The differences between the nominal and real azimuth antenna patterns for all swaths and polarisations for the SM, IW and EW modes.

2.2.2. Elevation Antenna Patterns (EAPs)

Figure 9 shows the comparison between the real and nominal elevation antenna patterns. The plots show there is not a significant change in the shape of the patterns, but there is a constant drop in power of around 0.6dB to 0.7dB over the main beam for the real patterns compared to the nominal patterns. Other plots for the Elevation antenna patterns can be found in RD-4.

The PG correction compensates for the drop of EAP power.




FO

RM

-NT-G

B-7

-1

Figure 9 Comparison between the nominal and real elevation antenna patterns. The plots on the left shows the real and nominal patterns for the HH polarisation. The plots

on the right show the difference between the real and nominal elevation antenna patterns for all polarisations.




FO

RM

-NT-G

B-7

-1

2.2.3. Steered AAP and descalloping function

Figure 10 and Figure 11 show the difference between azimuth patterns before and after tile 5 failure, for IW and EW modes. Patterns were calculated through S1 CFI AM, considering the EMs shown in Figure 1 and Figure 2. Difference has been plotted within -3 dB region. The observed difference for all beams is a ramp along azimuth, meaning that after tile failure the pointing of all beams slightly changed. This is also confirmed by the analysis of the Doppler Calibration Profile, represented in Figure 12 for the IW beams. The top image represents the DCP before tile failure and the bottom image represents the DCP after tile failure. A clear offset of the DCP (about 5 Hz) can be observed for all beams and polarization and the electronic mispointing of the antenna results slightly mitigated after tile failure. (Please ignore the spike at the centre of IW1 in the DCP as it is a known AM software issue).




FO

RM

-NT-G

B-7

-1

Figure 10: Difference between azimuth patterns of IW TopSAR beams for different ABT indexes.




FO

RM

-NT-G

B-7

-1

Figure 11: Difference between azimuth patterns of EW beams for different ABT indexes.




FO

RM

-NT-G

B-7

-1

Figure 12: Doppler Calibration Profile for IW beams before (top figure) and after (bottom figure) tile 5 failure.

Figure 13 and Figure 14 show the comparison between the currently used de-scalloping patterns for IW and EW modes (blue lines) and the newly calculated de-scalloping patterns exploiting S1 CFI AM. The agreement is very good and there is no need to modify the currently used de-scalloping functions.




FO

RM

-NT-G

B-7

-1

Figure 13: Comparison between currently used de-scalloping pattern (blue line) and de-scalloping pattern calculated with S1 CFI AM (red dashed line) for IW beams.




FO

RM

-NT-G

B-7

-1




FO

RM

-NT-G

B-7

-1

Figure 14: Comparison between currently used descalloping pattern (blue line) and descalloping pattern calculated with S1 CFI AM (red dashed line) for TopSAR EW beams

2.3. AAP phase ramp targets shifting effects

Figure 15 shows the AAP phase, before (top) and after (bottom) tile 5 failure, for Stripmap beams. The same trend can be observed for IW and EW beams as well and is originated by the antenna illumination broken-symmetry along azimuth. Such phase ramp, if not compensated during focusing can introduce a shift of the focused targets, whose magnitude can be derived from the well-known relationship between frequency phase ramp and delay. In particular a frequency phase ramp

tfje

2

introduces a time delay t . The value of the time delay can be obtained as:

vst

4

where s is the measured slope (50 deg/deg), is the sensor wavelength (0.055 m) and v is the

ground velocity (about 7000 m/s). The resulting delay is about 29 µs corresponding to approximately 0.2 m of targets shift in the focused images.




FO

RM

-NT-G

B-7

-1

Figure 15: AAP phase before (top) and after (bottom) tile 5 failure for Stripmap beams.

S1

S2

S3

S4

S5

S6

S1

S2

S3

S4

S5

S6




FO

RM

-NT-G

B-7

-1

3. Impact on L1 data products quality

3.1. EAP from AM: consistency with Gamma profiles

Figure 16 shows some examples of the Gamma profiles that have been calculated from L1 imagery acquired over the homogeneous Amazon rainforest since the TRM failures. In general the gamma profiles are relatively flat, indicating that the elevation antenna pattern has not significantly changed in shape due to the TRM failures. However, there are a few trends, for example, the gamma profile for IW3 VV shows a gradual increase, which does indicate a change in shape of the elevation antenna pattern. Other recent Amazon Gamma profiles can be found in RD-5.

Figure 16 Gamma profiles calculated from L1 products. Top left is S1, top right is S6, mid left is IW1, mid right is IW3, bottom left is EW1. The red and black lines indicate the

different polarisations.




FO

RM

-NT-G

B-7

-1

3.2. AAP from AM: whitening verification

Range compressed Stripmap data acquired over strong point targets such as transponders have been used to verify the data profile along azimuth (representing the 2-way AAP) against the AM prediction. Figure 17 below shows an example for the dataset S1A_S6_RAW__0SDH_20140628T052333_20140628T052403_001246_00133B_D923.SAFE that includes the Transponder D39 (over Ochsenhausen).

The AAP is estimated from the maximum intensity of the samples belonging to the transponder and is then superimposed to the AM prediction (see Figure 18). The comparison shows: a quite good fitting of the Antenna model pattern for both polarizations. In the main lobe part (around -3 dB beam-width) the pattern difference shows:

Co-pol: fluctuations around a constant value

Cross-pol: fluctuations with a residual azimuth varying behaviour

Similar conclusions can be drawn from the analysis of an IW image (S1A_IW_RAW__0SDV_20141010T053415_20141010T053447_002763_0031B2_954D over Transponder D41 - Bergerhausen). In this case the shrunk antenna pattern has been used as reference. The comparison is significant in the main-lobe only due to limited signal to clutter ratio for the side-lobes (Figure 19).

Figure 17: (a) Transponder D39 (Ochsenhausen) RGC image intensity. (b) Selection of 7 range samples and 1.2*PRF azimuth lines around the PT maximum. (c ) Estimated 2-w

Azimuth Antenna Pattern from data.

200 400 600 800 1000 1200 1400 1600 1800

1

2

3

4

5

6

7

-80

-70

-60

-50

-40

-30

-20

200 400 600 800 1000 1200 1400 1600

50

100

150

200

250

-90

-80

-70

-60

-50

-40

-30

-20

Rg

[dB]

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0Azimuth antenna pattern

Azimuth angles [deg]

[dB

]

[dB]

(a) (b) (c)




FO

RM

-NT-G

B-7

-1

Figure 18: Comparison between data profile (blue) and AAP from the AM, Stripmap S6 case.

Figure 19: Comparison between data profile (blue) and AAP from the AM, IW1 case.

3.3. Impact of antenna failures: update of the expected theoretical performance (NESZ and ambiguity)

The theoretical models for SAR performance prediction have been run with the following updates to configuration:

-antenna failure status at 20 January 2015

-azimuth processing bandwidth and Hamming window as in Table 2

-azimuth spectrum whitening active

-values computed at mid-burst for IW and EW modes

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3-40

-30

-20

-10

0

10

20

30

40Azimuth antenna pattern difference


[dB

]

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3-40

-30

-20

-10

0

10

20

30



[dB

]

(HH)

(HV)

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3-60

-50

-40

-30

-20

-10

0Azimuth antenna pattern comparison


[dB

]

Data pattern

AM pattern

-0.1 -0.08-0.06-0.04-0.02 0 0.02 0.04 0.06 0.08 0.1-3

-2.5

-2

-1.5

-1

-0.5

0

0.5



[dB

]

-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3-60

-50

-40

-30

-20

-10



[dB

]

Data pattern

AM pattern

-0.1 -0.08-0.06-0.04-0.02 0 0.02 0.04 0.06 0.08 0.1-3

-2.5

-2

-1.5

-1

-0.5

0

0.5



[dB

]

-0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-30

-20

-10

0

10

20

30

40

50

60



[dB

]

-0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-30

-20

-10

0

10

20

30

40

50

60



[dB

](VV)

(VH)

-0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-90

-80

-70

-60

-50

-40

-30

-20

-10



[dB

]

Data pattern

AM pattern

-0.03 -0.02 -0.01 0 0.01 0.02 0.03-7

-6

-5

-4

-3

-2

-1

0

1


Azimuth angles [deg][d

B]

-0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2-90

-80

-70

-60

-50

-40

-30

-20

-10



[dB

]

Data pattern

AM pattern

-0.04 -0.03 -0.02 -0.01 0 0.01 0.02-7

-6

-5

-4

-3

-2

-1

0

1



[dB

]




FO

RM

-NT-G

B-7

-1

S1 S2 S3 S4 S5 S6 IW1 IW2 IW3 EW1 EW2 EW3 EW4 EW5 WV1 WV2

Hamm 0,75 0,75 0,75 0,75 0,65 0,75 0,7 0,75 0,75 0,5 0,75 0,75 0,75 0,75 0,75 0,75

Band 1392 1405 1399 1429 1540 1398 327 313 314 233 159 156 154 159 1428 1429

Table 2: Azimuth processing band and Hamming coefficients considered for performance evaluation.

Figure 20 below reports the NESZ comparison no-failures vs. failures for all swaths. It can be noticed that the impact is a constant offset of about 0.6 dB for all cases.

Figure 20: Theoretical NESZ performance: comparison between expected values with no-failures (dashed line) and with (solid line) the antenna failures as in Figure 1.

Table 3 below reports the azimuth ambiguity to signal ratio comparison, failures vs. no-failures.

Az-DTAR [dB] S1 S2 S3 S4 S5 S6 IW1 IW2 IW3 EW1 EW2 EW3 EW4 EW5 WV1 WV2

no-failures -30,6 -21,9 -30,9 -22,9 -29,2 -24,9 -27,9 -27,1 -27,7 -32,3 -32,9 -32,2 -34,9 -32,2 -22,9 -22,9

failures as

in Figure 1 -27,8 -21,6 -28,3 -22,4 -27,1 -24,0 -25,9 -25,3 -25,8 -27,9 -30,2 -27,9 -30,8 -27,9 -22,4 -22,4

Table 3: Azimuth processing band and Hamming coefficients considered for performance evaluation.

15 20 25 30 35 40 45 50-30

-29

-28

-27

-26

-25

-24

-23

-22

-21Stripmap, altitude 697.74 Km

Incidence Angles [deg]

NE

SZ

[d

B]

S1

S2

S3

S4

S5

S6

30 32 34 36 38 40 42 44 46-31

-30

-29

-28

-27

-26

-25

-24

-23

-22InterferometricWideswath, altitude 697.74 Km


NE

SZ

[d

B]

IW1

IW2

IW3

15 20 25 30 35 40 45 50-36

-34

-32

-30

-28

-26

-24

-22ExtraWideswath, altitude 697.74 Km


NE

SZ

[d

B]

EW1

EW2

EW3

EW4

EW5

22 24 26 28 30 32 34 36 38-28

-27

-26

-25

-24

-23

-22

-21Wave, altitude 697.74 Km


NE

SZ

[d

B]

WV1

WV2




FO

RM

-NT-G

B-7

-1

It can be observed a degradation of the ambiguity rejection of [0.3 .. 2.8] dB for the Stripmap swaths, of [1.9 .. 2.0] dB for IW and [2.7 .. 4.4] dB for EW. For wave mode, the impact is the same of the Stripmap S4 case, since they share the same PRF. Despite the significant degradation of some cases, the ambiguity rejection remains better than -22 dB, except the only case of S2 that is -21.6 dB.

3.4. Impact of tile-5 intermittent power recovery on radiometric quality

It has been noticed that the recovery of power for the tile 5 can happen at any moment in the timeline. This is the case of for the DT 34BD(hex) acquired the 20/10 as illustrated in Figure 21. When this is happening, it can be seen that there are radiometric jumps from slice to slice likely related to wrong PG interpolation caused by the PG jump.

(a) PG evolution during DT 34BD (hex)

(b) radiometric jump associated to the recovery

Figure 21: Effect of intermittent recovery happening during a data-take




FO

RM

-NT-G

B-7

-1

3.5. TOPS De-scalloping verification

The failure of an entire tile would impact the shape of the azimuth antenna pattern. In the S-1 IPF there are two azimuth antenna patterns are used:

The classical 2-way azimuth antenna pattern used for whitening the spectra prior Hamming

windowing

The azimuth pattern of the antennas element used for TOPS de-scalloping

These 2 patterns are computed by the AM as shown in section 2.2.3.

Both corrections are impacting the radiometric normalisation. The fig below presents the results obtained for two EW product before/after the anomaly. EW has been chosen as it represents the worst-case scenario. As it can be seen from Figure 22, no major degradation is seen.

(a) EW1 VV after (left ) vs. before (right) the anomaly. It can be seen that a slight degradation is visible at the start of the burst where the measurement (black & blue) slightly deviates from the AM (pink)

(b) EW2 VV after (left ) vs. before (right) the anomaly. No visible change




FO

RM

-NT-G

B-7

-1

(c) EW3 VV after (left ) vs. before (right) the anomaly. No visible change

(d) EW4 VV after (left ) vs. before (right) the anomaly. No visible change

(d) EW5 VV after (left ) vs. before (right) the anomaly. No visible change

Figure 22: Verification of the TOPS de-scalloping correction made for 2 EW product after (left) and before (right) the anomaly.




FO

RM

-NT-G

B-7

-1

4. Conclusions

The evolution of the status of the antenna of the Sentinel-1A and the related analyses carried out at MPC side has been presented throughout the document.

The antenna TRM failures have impact on the antenna pattern shape in elevation and azimuth and on the antenna directivity.

The impact on the EAP is rather limited, and it is correctly modelled by the antenna model (it has been verified with the gamma profiles that there is no evident degradation due to the failures).

The AAP is impacted mainly in the side-lobes, causing a degradation of the azimuth ambiguity to signal ratio (DTAR) that ranges from 0.3 to 4.4 dB, depending on the acquisition mode and sub-swath.

The loss of directivity causes a bias that is captured by the PG and compensated for in the processor. Also, a NESZ degradation of 0.6 dB compared to the nominal case occurs.

A part from the degradation of the SAR performance (NESZ, DTAR), no other degradation on the data radiometric accuracy has to be reported except for some radiometric jumps on few slices, caused by the intermittent power recoveries of tile number 5.

The appearance of a phase ramp in the azimuth antenna pattern may cause an azimuth localization error in terms of a shift in the order of 0.2m for the Stripmap modes. The value has been predicted with theoretical calculations, but not yet verified from data.

Basing on the above analyses and on the current knowledge, we conclude that there is no major degradation on the data in keeping the current instrument configuration.




FO

RM

-NT-G

B-7

-1

Appendix A - List of acronyms

TBC To be confirmed

TBD To be defined

AD Applicable Document

RD Reference Document

NESZ Noise Equivalent Sigma Zero

DTAR Distributed Target Ambiguity Ratio

AM Antenna Model

EAP Elevation Antenna Pattern

AAP Azimuth Antenna Pattern

TRM Transmit-Receive Module

quality disclaimersentinel -1 no. 02...sentinel-1a tile #5 intermittent failure: impact on l1...

Documents