[SLS-DC] FW: [EXTERNAL] WG: CCSDS DC - SAR compression meeting summary: action item updates

Wed Sep 6 13:15:24 UTC 2023

Mark Wong
NASA GSFC Flight Microwave & Telecommunication Systems Branch
Phone: 1-301-286-6549

From: Michele.Martone at dlr.de <Michele.Martone at dlr.de>
Sent: Wednesday, September 6, 2023 8:12 AM
To: sls-dc-bounces at mailman.ccsds.org
Cc: Wong, Englin (GSFC-5670) <mark.wong at nasa.gov>
Subject: [EXTERNAL] WG: CCSDS DC - SAR compression meeting summary: action item updates

CAUTION: This email originated from outside of NASA.  Please take care when clicking links or opening attachments.  Use the "Report Message" button to report suspicious messages to the NASA SOC.

Dear colleagues,

I just wanted to give you a quick update on some of the Actions agreed at the last CCSDS meeting.

  1.  DLR will provide a reference list to FDBAQ for the working group: yesterday Lucana and I had a call on this topic, we identified a list of publications and project reports/TN which can be used for a better understanding of the FDBAQ algorithm. Lucana is going to check the availability of the ESA documentation and let us know.
  2.  DLR will also provide a potential min/max SQNR range for the SAR instrument. This may depend on the input bit-width: we have conducted some simulations of a complex signal with normally distributed Re and Im quantized with ADC at 8 bps, 12 bps and 16 bps, the latter representing an adequate reference for SAR instrument as agreed at the last meeting (see comment 1. here below). The plot is here attached, and shows the mean SQNR as function of the signal-to-clipping ratio (gamma_clip), this parameter being related to the ratio of the input signal power and the quantizer clipping value. Too small values of gamma_clip (left part of the plot) result in increased granular noise, while large gamma_clip values (right part) determine dominant clipping errors, both cases resulting in a degraded SQNR. The highest mean SQNR is ~ 90 dB and the stddev ~ 13 dB. According to this, an SQNR range between 0 and 110 dB is in our opinion realistic and representative for SAR instruments.
Should you have questions please just let me know.

Thank you and best regards,
Michele

--------------------------
Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR)
Institut für Hochfrequenztechnik und Radarsysteme | Oberpfaffenhofen | 82234 Weßling

Dr. Michele Martone | Satelliten-SAR-Systeme
Telefon 08153 28-1286 | Telefax 08153 28-1449 | Michele.Martone at dlr.de<mailto:Michele.Martone at dlr.de>
www.DLR.de<http://www.dlr.de/>

Von: SLS-DC <sls-dc-bounces at mailman.ccsds.org<mailto:sls-dc-bounces at mailman.ccsds.org>> Im Auftrag von Wong, Englin (GSFC-5670) via SLS-DC
Gesendet: Dienstag, 20. Juni 2023 14:37
An: sls-dc at mailman.ccsds.org<mailto:sls-dc at mailman.ccsds.org>
Betreff: [SLS-DC] CCSDS DC - SAR compression meeting summary: action item updates

Here are the updated action items for the SAR compression based on feedback from DLR.

What the working group agrees on:

  1.  The 16-bit I and 16-bit Q inputs to FDBAQ are adequate for the current SAR instrument. The 16-bit input provides 80 dB of dynamic range, which is plenty.
  2.  The working group concluded that to minimize the ADC description, a more focused description of the input data format is needed. Currently, there is a description of integer input, but fractional equations are present in the white book.
  3.  UAB agrees to support DLR in the next iteration of the white book. This is an attempt to move the white book towards a more CCSDS-compliant write-up.

Open questions:

  1.  The definition of the input mid-raised quantizer is inherited from BAQ. However, it is unclear if this is the optimal solution. An investigation into the BAQ might be needed.
  2.  Multiple LUTs described in the white book refer to optimization based on SAR instrument calibration. It is unclear how these steps are performed. The working group agrees that optimization is important for users to recreate these tables. Questions regarding additional inputs such as thermal noise, SQNR, and compression rate could potentially be defined as input parameters were raised.
  3.  The discussion also touched on Huffman coding table and discuss the generation of this table is also based on the optimization. The optimization is based on SQNR and compression rate.

Action Items:

  1.  ESA to upload the DLR white book response excel sheet to share drive.
  2.  DLR will provide a reference list to FDBAQ for the working group. (DLR: by 8/15/2023)
  3.  DLR will also provide a potential min/max SQNR range for the SAR instrument. This may depend on the input bit-width. (DLR: During Fall meeting)
  4.  DLR will review the full FDBAQ documentation (including the TNs to be delivered by ESA) and identify the key aspects to be included in the description of the standard. (DLR: by Fall meeting)
  5.  The working group will provide written feedback to the current white book draft one month before the next meeting (ALL: By 10/5/2023).

Best,
Mark

---------------------------------------------------------------------------------------
Mark Wong, Ph.D.
NASA/Goddard Space Flight Center
Flight Microwave & Telecommunication Systems Branch/Code 567
Greenbelt, MD 20771
B19 Rm S030
Phone: 1-301-286-6549
----------------------------------------------------------------------------------------

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ccsds.org/pipermail/sls-dc/attachments/20230906/71a70906/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SQNR_ADC.png
Type: image/png
Size: 92018 bytes
Desc: SQNR_ADC.png
URL: <http://mailman.ccsds.org/pipermail/sls-dc/attachments/20230906/71a70906/attachment-0001.png>