Announcement

Collapse
No announcement yet.

Archived - Beta test MLAT software for Raspberry Pi

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Archived - Beta test MLAT software for Raspberry Pi

    We are happy to announce public beta testing phase of MLAT based on Raspberry Pi receivers.

    If you would like to participate, please make sure you have the following:
    • Raspberry Pi B+ or Raspberry Pi 2
    • DVB-T USB receiver with antenna
    • Fairly good location to install the antenna


    Local tests were performed on RPi B+ and RPi2 with Raspbian OS, but it still may work on older models and/or other OS.

    Current plan is to provide the decoder software through our APT repository in order to simplify update process.

    Updated Dec 2015
    Automated install script:

    Code:
    $ sudo bash -c "$(wget -O - http://repo.feed.flightradar24.com/install_fr24_rpi.sh)"
    Original (now outdated) Manual method:

    First you will need to configure the repository, importing our keys and adding the repository URL:
    Code:
    $ gpg --keyserver pgp.mit.edu --recv-keys 40C430F5
    $ gpg --armor --export 40C430F5 | sudo apt-key add -
    $ sudo nano /etc/apt/sources.list
    Add this line to /etc/apt/sources.list:
    Code:
    deb http://repo.feed.flightradar24.com flightradar24 raspberrypi-beta
    Then update the cache and just run regular install command:
    Code:
    $ sudo apt-get update
    $ sudo apt-get install fr24feed
    The package provides everything required to run the decoder software, including modified dump1090 utility which is forked from Malcolm Robb's branch but has several improvements for FR24 MLAT. It may be incompatible with other MLAT software if you use some.

    After installation, please run the sign-up wizard even if you already have the sharing key:
    Code:
    $ fr24feed --signup
    You will be asked for a sharing key, whether or not you want MLAT enabled and for your exact position. Please carefully input the antenna's position, including altitude, because it's important for MLAT calculations. Other sign-up steps remain the same as before.

    After sign-up process is complete, you may start feeding data by issuing:
    Code:
    $ sudo service fr24feed restart
    As usual you may monitor the app status by:
    Code:
    $ sudo fr24feed-status
    [ ok ] FR24 Feeder/Decoder Process: running.
    [ ok ] FR24 Stats Timestamp: 2015-07-01 07:55:45.
    [ ok ] FR24 Link: connected [UDP].
    [ ok ] FR24 Radar: T-XXXX12.
    [ ok ] FR24 Tracked AC: 2.
    [ ok ] Receiver: connected (3313 MSGS/0 SYNC).
    [ ok ] FR24 MLAT: ok [UDP].
    [ ok ] FR24 MLAT AC seen: 5.
    Please note that you won't see the immediate result on FR24 map. The reason for that is that we want first to evaluate the data and configure our smoothing filters appropriately as normal errors for RPi MLAT can easily be around 10km (this is due to non precise system clock of RPi). That will take some time, but we will all benefit from having smooth trails instead of ugly jumps.

    Thanks!

    UPD: Discussion on how to run fr24feed with other feeding software
    Last edited by Oblivian; 2019-04-13, 09:51. Reason: Change status command

    #2
    Oh nice!

    That should spread the cover a little more in built up areas
    Posts not to be taken as official support representation - Just a helpful uploader who tinkers

    Comment


      #3
      Great Stuff, Software Develoment team. Lets hope its going to better MLAT even more.

      But I have to ask: when can we expect a Linux build?

      PS: About the time drift/time offset: why not build a NTP client into the software (or load it additional to the system) and force the RasPi to update its time every few hours (like done on PlanePlotter)
      Last edited by HermanZA; 2015-07-01, 09:07.

      Comment


        #4
        is this also available for normal linux servers?

        Comment


          #5
          We start with just Raspberry Pi and when it has been tested we will add more platforms.

          Comment


            #6
            There is NTP synchronization. The problem is that's it's not enough for MLAT. When we speak in terms of speed of light, 100 microseconds error gives 30km difference.

            Comment


              #7
              Does MLAT data from RPI is used in conjuction with radarcape sources or mlat positions are generated only between T- feeders?

              Comment


                #8
                Originally posted by giacomo892 View Post
                Does MLAT data from RPI is used in conjuction with radarcape sources or mlat positions are generated only between T- feeders?
                It's used together with other MLAT sources, but...

                Please note that you won't see the immediate result on FR24 map. The reason for that is that we want first to evaluate the data and configure our smoothing filters appropriately as normal errors for RPi MLAT can easily be around 10km (this due to non precise system clock of RPi). That will take some time, but we will all benefit from having smooth trails instead of ugly jumps.

                Comment


                  #9
                  Originally posted by Mike View Post
                  It's used together with other MLAT sources, but...
                  Thanks Mike. I'm already feeding MLAT with RPI.

                  Is there a way to access early data?

                  Comment


                    #10
                    It's working:
                    Code:
                    pi@raspberrypi ~ $ sudo service fr24feed status
                    [ ok ] FR24 Feeder/Decoder Process: running.
                    [ ok ] FR24 Stats Timestamp: 2015-07-01 10:14:09.
                    [ ok ] FR24 Link: connected [UDP].
                    [ ok ] FR24 Radar: T-EHTW3.
                    [ ok ] FR24 Tracked AC: 163.
                    [ ok ] Receiver: connected (4050312 MSGS/0 SYNC).
                    [ ok ] FR24 MLAT: ok [UDP].
                    [ ok ] FR24 MLAT AC seen: 166.
                    Last edited by T-EHTW3; 2015-07-01, 10:19.
                    http://www.twenteradar.nl

                    Comment


                      #11
                      Can you document the changes to the timestamp format that you have made, please?

                      (I'm not sure why it needed to change, the 12MHz clock works fine, but whatever the reason I'm inevitably going to be asked to support this in mlat-client at some point)

                      edit: possibly assuming too much here. You say there are changes to dump1090 that may be incompatible with other mlat clients. What are those changes?
                      editedit: oh! source on github. Let me have a look..
                      edit^3: OK, so it'll break anything that wants to use AVR output, but the Beast output is unmodified.
                      Last edited by obj; 2015-07-01, 16:20.

                      Comment


                        #12
                        runining fine

                        pi@monitoring /etc $ sudo service fr24feed status
                        [ ok ] FR24 Feeder/Decoder Process: running.
                        [ ok ] FR24 Stats Timestamp: 2015-07-01 14:54:27.
                        [ ok ] FR24 Link: connected [UDP].
                        [ ok ] FR24 Radar: T-EHSB67.
                        [ ok ] FR24 Tracked AC: 124.
                        [ ok ] Receiver: connected (764002 MSGS/0 SYNC).
                        [ ok ] FR24 MLAT: ok [UDP].
                        [ ok ] FR24 MLAT AC seen: 132.

                        also running gps / ntp

                        pi@monitoring /etc $ ntpq
                        ntpq> pe
                        remote refid st t when poll reach delay offset jitter
                        ================================================== ============================
                        oGPS_NMEA(0) .GPS. 0 l 4 16 377 0.000 0.000 0.002
                        +yao.tokyobase.n 213.75.85.246 2 u 49 64 377 7.067 -1.908 0.153
                        *auth1.xs4all.nl 193.79.237.14 2 u 47 64 377 5.756 0.177 0.037

                        Comment


                          #13
                          After I install it on my raspberry pi B+ ,

                          pi@raspiblack ~ $ service fr24feed status
                          [ ok ] FR24 Feeder/Decoder Process: running.
                          [ ok ] FR24 Stats Timestamp: 2015-07-01 16:17:37.
                          [ ok ] FR24 Link: connected [UDP].
                          [ ok ] FR24 Radar: T-ZSSS32.
                          [ ok ] FR24 Tracked AC:.
                          [FAIL] Receiver: down ... failed!
                          [FAIL] FR24 MLAT: not running ... failed!


                          So I check the web:8754 status

                          the version show Linux/generic/armv7l/1.0.13-6

                          not armv6 ? ?
                          T-ZSSS32 F-ZSSS1 force shutdown by local regulation

                          Comment


                            #14
                            Originally posted by Mike View Post
                            We start with just Raspberry Pi and when it has been tested we will add more platforms.
                            Thanks Mike. Like many others, I too am anxiously awaiting the Linux update!

                            Rick
                            T-CYWH3

                            F-CYWH1 Southern tip of Vancouver Island

                            Comment


                              #15
                              Originally posted by vinnyspb View Post
                              ...
                              You will be asked for a sharing key, whether or not you want MLAT enabled and for your exact position. Please carefully input the antenna's position, including altitude, because it's important for MLAT calculations. Other sign-up steps remain the same as before.

                              ...
                              Thanks!

                              What is the best way to determine the exact position (including altitude)? And how exact should it be?

                              Is there a possibility to get the exact position via iPhone GPS?

                              Comment

                              Working...
                              X