2018-08-22 - Primary Payload discussion - Meeting notes

Date

22 Aug 2018

Attendees

• Unknown User (sivertba)

• Unknown User (larshb)

• Unknown User (santora)

Goals

• Better define what the payload should do
• Better define how the payload should do the things it is supposed to do

Notes

• There are still a lot of confusion and questions regarding the Hardware interface of the payload
• clear benefit between multispectral and hyperspectral should be defined
  • benefits of multispectral
    • This is how the spectral resolution of most OC satellites
      
      • algorithms and models are based and tested on this spectral resolution
    • We will achieve a better signal to noise ratio using bands i.e. binning more signals together
    • The HABsat-1 mission claim: "Research has shown that hyperspectral custom narrowband multispectral imagers with radiometric thermal imagers are the preferred instruments for predicting, detecting, identifying potential HABs in the oceans"
      • https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=4250&context=smallsat
      • https://digitalcommons.usu.edu/smallsat/2018/all2018/438/
    • Will produce less data, alleviating the burden wrt. storage requirements and downlink speed
  • benefits of hyperspectral
    
    • The mission is supposed to be hyperspectral
    • We can take a more statistical approach as opposed to relying on "perfect" bands
    • "To identify phytoplankton functional groups, hyperspectral data with 5 nm resolution is required"
      • http://ioccg.org/wp-content/uploads/2015/10/ioccg-report-13.pdf
    • "Research has shown that hyperspectral or custom narrowband multispectral imagers with radiometric thermal imagers are the preferred instruments for predicting, detecting, identifying potential HABs in the oceans"
      • https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=4250&context=smallsat
    • Greater flexibility wrt. to potential changes in operation
• There is a clear distinction between classification and target detection. Both are applications within hyperspectral image processing, but not dependent on each other.
  • Could use classification schemes to determine regions of interest, but at an undetermined power/computational cost. 
• Even though not clarified in the meeting the CCSDS-123 is the better choice for lossless compression given TRL for our mission when compared to JPEG2000
  • https://drive.google.com/open?id=1T3ZnOqW13oscD0ZXk7_cZrwRJ_52nmrN
  • based on comments from Unknown User (larshb) and Unknown User (orlandic)

Action items

• Determine a regular weekly or bi-weekly meeting schedule Unknown User (sivertba)
• Share the pipeline draft Unknown User (santora)
• Share the first draft of interfacing for primary payload Unknown User (larshb)
• Comment on the Pipeline draft made by Mariusz Unknown User (sivertba)