Announcement

Collapse
No announcement yet.

Blue satellite tracks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Blue satellite tracks

    Hi, nice to see that the blue satellite-tracked flights are now available!
    It did raise some questions, maybe someone can elucidate?

    Blue flight make a dotted black track as if it is out of coverage. Normally it is a color coded track showing altitude. Is altitude data also available from the satellite feed, and will this be used for the track information in the future?

    Why are some oceanic flights (middle of nowhere) yellow (estimated) while a close by flight is blue (satellite). I would think both signals should be received by an overhead satellite?

    Your information cites some satellite sources such as Spire, but not Aireon which is used by Flightaware. Is this indeed not available to you? What is the coverage of the constellation that you can use? (i.e. almost continuous or occasional updates when a satellite happens to pass)

    I noticed satellite tracks are missing from playback, will this be implemented at some point?

    Just curious, thanks,

    Martijn

  • #2
    I'm not convinced the blue iconed flights are ADS-B positions tracked by satellite. I could be wrong but......

    I have the satellite option turned off and the tracks are usually black dotted lines, great circle curves from last terrestrial ADS-B position to destination. They appear to me to be the old estimated tracks now highlighted with blue icons.
    Mike


    www.radarspotting.com

    Radarspotting since 2005

    Comment


    • #3
      I believe flight radar blog tells you more but the upload is only every 15 or 20 minutes so it doesn't know the bits in between a has to estermate
      Posts not to be taken as official support representation - Just a helpful uploader who tinkers

      Comment


      • #4
        Originally posted by Oblivian View Post
        I believe flight radar blog tells you more but the upload is only every 15 or 20 minutes so it doesn't know the bits in between a has to estermate
        The blog says: "Due to the possible latency of up to 10 minutes depending on the satellite source..."
        which sort of suggests it may be up to 10 min until the next satellite get into range to collect one or more datapoints, after which it gets out of range again and we wait up to 10 min when the next one gets into range.

        In this way, the system gets at dotted track of the flight in and out of range. In spite of this limitation, I think it would be nice to plot that track data.

        Comment


        • #5
          I'm not quite sure I understand, but you can't plot something you don't have

          Spires system themselves is described as only sampling altitude, lat, long every 15mins to meet ICAO mandate. Vs standard ADSB of every 2 seconds. You can cover a lot of ground in that space so expect speed and so on is derived from pt to point average (look at skewed MLAT results for the downside of trying to plot this)

          It looks by the example I just viewed this is what is happening. Ground speed only.. and it was spinning around a number of degrees on the spot

          Dot/few position updates to help advance where one is vs guessing between the wider space it had to previously and throwing 'estimations' way out.
          But as there is no breadcrumb trail it looks as if FR24 is simply using these infrequent updates to improve 'estimations'

          Plotting the entire path with the raw data over and above the dotted 'this is the standard route' looks like it would falsely cause odd detours and kinks in an otherwise straight route track.

          Non sat A320 ​​airframe, shown but 'estimated'

          satfr24.JPG
          SAT example I looked at, with a kink already present and falsely advancing (icon soon started to turn all over the joint)
          sqsat.JPG
          Attached Files
          Last edited by Oblivian; 2020-03-16, 01:21.
          Posts not to be taken as official support representation - Just a helpful uploader who tinkers

          Comment


          • #6

            From a system perspective there is hardly a difference between sampling position every 2 seconds (as broadcast), every minute (as stored by FR24 in csv logs) or every 10 min (~as received from sat). There is always some interpolation/extrapolation within the limits of the granularity of the underlying data .

            I made 3 screenshots of a flight over the Atlantic, you can see the track of the (blue) flight goes south over time, suggesting that for now they do not plot the known track based on available sat breadcrumbs, but the great circle connecting current (sat) position and the last known terrestrial ADSB.



            I did see another flight reporting more erratic jumping positions, quite possible some robustness is lacking for now, and that is why they chose for this strategy for now.


            BlueTracks.png

            Jumps.png

            Comment


            • #7
              FYI, I looked up the Spire satellite constellation, one of the data providers.This gives a sense of their coverage. Their Lemur 2 sats are ADS-B capable starting late 2018 launches, that is about 1/3 of those selected in the picture. Also polar orbits included among them for high latitudes.

              Lemur.png

              Comment


              • #8
                It's always shown great circle between adsb points. Even before sat updates. (That's kinda what I was trying to get at). Between that It was a best guess and often went off course at the end/start like your 2nd image transition or had bad ETA or 'estimated' position (if you turn that on, I have it off to stop over-runs and more accurate speed change)

                Similar to the MLAT and flarm addition it seems to be just something FR have thrown in to try and keep ahead of the game or improve on their best-guesses implemented stuff.

                Sat updates seem to do not a lot but confirm the current position somewhere between those 2 adsb points, or give an updated shorter guessed curve/dashed 'probably somewhere here' line (vs ideally log the last few thousand position locations and supply during an upload/pass to back-date the map) so only improvement seems to be a sharper avg speed update and or time to arrival(till in-adsb) update
                Posts not to be taken as official support representation - Just a helpful uploader who tinkers

                Comment


                • #9
                  On a different note (but this seemed the right spot to post), why is it that there are no, or virtually no satellite data points over Russia or China, despite both countries having huge uncovered. There seem to be quite a few over Africa. It's hard to see how this could be a government enforced situation, as they do not control the satellites. Can anyone clue me in please?

                  Comment

                  Working...
                  X