Skip to content
Branch: master
Find file Copy path
Find file Copy path
7 contributors

Users who have contributed to this file

@ahupowerdns @mahtin @philcrump @kiwibrew @wojas @CTassisF @alimirjamali
269 lines (210 sloc) 9.39 KB


galileo/GPS/GLONASS/BeiDou open source monitoring. GPL3 licensed. (C) AHU Holding BV - -

Live website:

Theoretically multi-vendor, although currently only the U-blox 8 and 9 chipsets are supported. Navilock NL-8012U receiver works really well, as does the U-blox evaluation kit for the 8MT. In addition, many stations have reported success with this very cheap AliExpress sourced device. The best and most high-end receiver, which does all bands, all the time, is the Ublox F9P, several of us use the ArdusimpleRTK2B board.


  • Processes raw frames/strings/words from GPS, GLONASS, BeiDou and Galileo
  • All-band support (E1, E5b, B1I, B2I, Glonass L1, Glonass L2, GPS L1C/A) so far, GPS L2C and Galileo E5a pending).
  • Calculate ephemeris positions
  • Comparison of ephemerides to independent SP3 data to determine SISE
    • Globally, locally, worst user location
  • Record discontinuities between subsequent ephemerides (in time and space)
  • Compare doppler shift as reported by receiver with that expected from ephemeris
  • Track atomic clock & report jumps
  • Coverage maps (number of satellites >5, >10, >20 elevation)
  • Compare orbit to TLE, match up to best matching satellite
  • Tear out every bit that tells us how well an SV is doing
  • Full almanac processing to see what should be transmitting
  • Distributed receivers, combined into a single source of all messages
  • Ready to detect/report spoofing/jamming

Data is made available as JSON, as a user-friendly website and as a time-series database. This time-series database is easily mated to the industry standard Matplotlib/Pandas/Jupyter combination.

There is also tooling to extract raw frames/strings/words from specific timeframes.


  1. Support multiple wildly distributed receivers
  2. Combine these into a forensic archive of all Galileo/GPS NAV messages
  3. Make this archive available, offline and as a stream
  4. Consume this stream and turn it into an attractive live website ( As part of this, perform higher-level calculations to determine ephemeris discontinuities, live gst/gps/galileo time offsets, atomic clock jumps etc.
  5. Populate an InfluxDB timeseries database with raw measurements and higher order calculations

Works on Linux (including Raspbian Buster on Pi Zero W), OSX and OpenBSD.

Build locally

To get started, make sure you have a C++17 compiler (like g++ 8 or higher), git, protobuf-compiler. Then run 'make ubxtool navdump' to build the receiver-only tools.

To build everything, including the webserver, try:

apt-get install protobuf-compiler libh2o-dev libcurl4-openssl-dev libssl-dev libprotobuf-dev \ 
libh2o-evloop-dev libwslay-dev libncurses5-dev libeigen3-dev
git clone --recursive
cd galmon

If this doesn't succeed with an error about h2o, make sure you have this library installed. If you get an error about 'wslay', do the following, and run make again:

echo WSLAY=-lwslay > Makefile.local

Build in Docker

To build it in Docker:

git clone --recursive
docker build -t galmon --build-arg MAKE_FLAGS=-j2 .

To run a container with a shell in there:

docker run -it --rm galmon


Once compiled, run for example ./ubxtool --wait --port /dev/ttyACM0 --station 1 --stdout | ./navparse html null

Next up, browse to http://[::1]:10000 (or try http://localhost:10000/ and you should be in business. ubxtool changes (non-permanently) the configuration of your u-blox receiver so it emits the required frames for GPS and Galileo. If you have a u-blox timing receiver it will also enable the doppler frames.

By default the ublox receiver module will be configured to use the USB port, if you want to use a different interface port on the ublox module then add the --ubxport <id> option using one of the following numeric IDs:

0 : DDC (aka. I2C)
1 : UART[1]
2 : UART2
3 : USB (default)
4 : SPI

To see what is going on, try:

./ubxtool --wait --port /dev/ttyACM0 --station 1 --stdout | ./navdump

To distribute data to a remote navrecv, use:

./ubxtool --wait --port /dev/ttyACM0 --station 255 --dest

This will send protobuf to You can add as many destinations as you want, they will buffer and automatically reconnect. To also send data to stdout, add --stdout.


  • ubxtool: can configure a u-blox 8 chipset, parses its output & will convert it into a protbuf stream of GNSS NAV frames + metadata Adds 64-bit timestamps plus origin information to each message
  • xtool: if you have another chipset, build something that extracts NAV frames & metadata. Not done yet.
  • navrecv: receives GNSS NAV frames and stores them on disk, split out per sender.
  • navnexus: tails the files stored by navrecv, makes them available over TCP
  • navparse: consumes these ordered nav updates for a nice website and puts "ready to graph" data in influxdb - this is the first step that breaks "store everything in native format". Also does computations on ephemerides.
  • grafana dashboard: makes pretty graphs

Linux Systemd

This is very much a first stab. Do the following at root.

mkdir /run/ubxtool
mkdir /usr/local/ubxtool
cp /usr/local/ubxtool/
chmod +x /usr/local/ubxtool/
cp ubxtool /usr/local/ubxtool/
cp ubxtool.service /etc/systemd/system/ubxtool.service
touch /usr/local/ubxtool/destination
touch /usr/local/ubxtool/station

Then edit /usr/local/ubxtool/destination with an IP address collected from Bert. Then edit /usr/local/ubxtool/station with a station number collected from Bert.

The start up the service.

sudo systemctl enable ubxtool
sudo systemctl start ubxtool

This will be cleaned up and better packaged sometime soon.

Distributed setup

Run navrecv :: ./storage to receive frames on port 29603 of ::, aka all your IPv6 addresses (and IPv4 too on Linux). This allows anyone to send you frames, so be aware.

Next up, run navnexus ./storage ::, which will serve your recorded data from port 29601. It will merge messages coming in from all sources and serve them in time order.

Finally, you can do nv 29601 | ./navdump, which will give you all messages over the past 24 hours, and stream you more. This also works for navparse for the pretty website and influx storage, nc 29601 | ./navparse html galileo, if you have an influxdb running on localhost with a galileo database in there.


The transport format consists of repeats of:

  1. Four byte magic value
  2. Two-byte frame length
  3. A protobuf frame

The magic value is there to help us resync from partially written data.

The whole goal is that we can continue to rebuild the database by rerunning 'navstore' and 'navinflux'.


Data sources

The software can interpret SP3 files, good sources:


  • Dual goals: completeness, liveness, not the same For forensics, great if the packet is there For display, not that bad if we missed a message

  • In general, consider refeed strategy Raw serial Protobuf Influxdb ".csv files"

  • Delivery needs to be bit more stateful (queue)

  • Semantics definition for output of Navnexus "we'll never surprise you with old data"

Global coverage (via volunteers)

In alphabetical order:

  • Austria (Vienna area)
  • Brazil
  • Holland (Nootdorp, Hilversum, etc)
  • India (New Delhi area)
  • New Zealand (Auckland area)
  • Rusia (Moscow area)
  • Singapore
  • South Africa (Cape Town area)
  • Spain
  • Tonga
  • USA California (Santa Cruz, Los Angeles area, etc)
  • Uruguay

Additional sites are welcome (and encouraged) as the more data receiving sites that exist, then more accurate data and absolute coverage of each constellation can be had.

The galmon project is very grateful to all it's volunteering receiving stations.


  • Will also spool raw serial data to disk (in a filename that includes the start date)
  • Can also read from disk
  • Careful to add the right timestamps
You can’t perform that action at this time.