Skip to content

dgiardini/rtl-ais

 
 

Repository files navigation

rtl-ais, a simple AIS tuner and generic dual-frequency FM demodulator

rtl-ais provides the rtl_ais command, which decodes AIS data from Software Defined Radio (SDR) and outputs AIVDM / AIVDO sentences.

OS support
Linux
Windows
OSX

Command Line

Use: rtl_ais [options] [outputfile]
        [-l left_frequency (default: 161.975M)]
        [-r right_frequency (default: 162.025M)]
            left freq < right freq
            frequencies must be within 1.2MHz
        [-s sample_rate (default: 24k)]
            maximum value, might be down to 12k
        [-o output_rate (default: 48k)]
            must be equal or greater than twice -s value
        [-E toggle edge tuning (default: off)]
        [-D toggle DC filter (default: on)]
        [-d device_index (default: 0)]
        [-g tuner_gain (default: automatic)]
        [-p ppm_error (default: 0)]
        [-R enable RTL chip AGC (default: off)]
        [-A turn off built-in AIS decoder (default: on)]
            use this option to output samples to file or stdout.
        Built-in AIS decoder options:
        [-h host (default: 127.0.0.1)]
        [-P port (default: 10110)]
        [-T use TCP communication as tcp listener ( -h is ignored)]
        [-t time to keep ais messages in sec, using tcp listener (default: 15)]
        [-n log NMEA sentences to console (stderr) (default off)]
        [-L log sound levels to console (stderr) (default off)]
        [-I add sample index to NMEA mesages (default off)]
        [-S seconds_for_decoder_stats (default 0=off)]
        When the built-in AIS decoder is disabled the samples are sent to
        to [outputfile] (a '-' dumps samples to stdout)
            omitting the filename also uses stdout
        Output is stereo 2x16 bit signed ints
        Examples:
        Receive AIS traffic,sent UDP NMEA sentences to 127.0.0.1 port 10110
             and log the senteces to console:
        rtl_ais -n
        Tune two fm stations and play one on each channel:
        rtl_ais -l233.15M  -r233.20M -A  | play -r48k -traw -es -b16 -c2 -V1 -

Compiling

Make sure you have the following dependencies:

  • librtlsdr
  • libusb
  • libpthread
$ # Get the source code:
$ git clone https://github.com/dgiardini/rtl-ais
$ # Change to the source dir
$ cd rtl-ais
$ make
$ # Test running the command
$ ./rtl_ais

For compiling a MS Windows executable you will need a working MSYS/MinGW environment. Edit the Makefile, and modify these lines:

#### point this to your correct path ###
RTLSDR_PATH="/c/tmp/rtl-sdr/"
RTLSDR_LIB=$(RTLSDR_PATH)/build/src/
########################################

Installing

  • On Linux, sudo make install
  • On Windows, put the librtlsdr.dll and libusb-1.0.dll files in the same directory with rtl_ais.exe. You'll need the zadig driver installed too.

Running

rtl-ais uses software defined radio (SDR). The specific hardware we use for this is a DVB-T dongle. A good starting point is: https://www.rtl-sdr.com/about-rtl-sdr

You'll need also an antenna, and be located near (some miles) the passing vessels.

You'll also need to do some procedure to get the tunning error for the specfic dongle you have (aka ppm error), and pass that number as parameter of rtl-ais.

Docker Container

Now you can run rtl-ais in a docker container. No dependencies to install. Total container size is approximately 75 MB. Get/install docker here.

Two options for obtaining the container: Either download and run a pre-built container, or build the container locally.

  1. Just to test things out: docker run -it --rm --device=/dev/bus/usb ghcr.io/bklofas/rtl-ais:latest

    • This downloads a pre-built container from the Github container registry.
    • This image will run by default ./rtl_ais -n, showing the received packets on STDOUT. All other default values.
    • You can add other ./rtl-sdr options, see below.
    • Make sure at least one RTL-SDR dongle is connected.
    • Startup messages and decoded packets will display in the terminal.
    • Ctrl-C to kill.
    • Using the --rm flag will delete the container when you kill it. Otherwise, it will stay around until you prune.
  2. For a more permanent setup, run the container in the background and add any options you want: docker run -d --name rtl-ais --restart=unless-stopped --log-driver=local --network=host --device=/dev/bus/usb ghcr.io/bklofas/rtl-ais:latest ./rtl_ais -n -d 00000002 -h 127.0.0.1 -P 10110

    • -d: Start this container in daemon/background mode.
    • --name: Name this anything you want.
    • --restart=unless-stopped: Automatically restart the container if something happens (reboot, USB problem), unless you have manually stopped the container (with docker stop rtl-ais).
    • --log-driver=local: By default, docker uses the json log driver which may fill up your harddrive, depending on how busy your station is. local log driver defaults to 100MB of saved logs, and automatically rotates them.
    • --network=host: Allows the container to talk to the internet, if you are sending the packets to an online service.
    • --device=: Allows the container to talk to the USB bus to access the RTL-SDR dongle.
    • ./rtl_ais: Same command-line options as above.
    • View the startup messages and decoded packets with docker logs --follow rtl-ais

Building the container:

  • git clone https://github.com/bklofas/rtl-ais.git the repository, then from the folder docker build -t rtl-ais .
  • Or, build the container without cloning the repository: docker build https://github.com/bklofas/rtl-ais.git

Other tips and tricks:

  • If you have the -n flag, view the decoded AIS packets in real-time with docker logs --follow rtl-ais
  • If you are only sending packets to one internet service (such as marinetraffic.com), you can use the -h and -P options that they send you.

Testing

TODO: something like https://github.com/freerange/ais-on-sdr/wiki/Testing-AISDecoder#with-an-audio-file

Known Issues

About

A simple AIS tuner and generic dual-frequency FM demodulator

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • C 80.1%
  • Python 17.5%
  • Makefile 1.9%
  • Dockerfile 0.5%