Foreword

This project is inspired and reuse several parts of several other ADS-B / mode S projects amongst which:

• https://github.com/lyusupov/ADSB-Out
• https://github.com/nzkarit/ADSB-Out and https://github.com/pynstrom/adsb-out
• https://github.com/bistromath/gr-air-modes
• https://github.com/junzis/pyModeS

All those repositories are published under GNU General Public License v3.0. This is also the license chosen for this repository.
Please let me know if you have issues or require more explicit citations about reused source code.

Project goals

The initial project goals are oriented towards:

• completing the set of broadcastable messages that have already been implemented “adsb-out” in referenced projects.
• fixing bugs / adding features in existing code.
• producing a software architecture that better suit my understanding/habits.
• beeing able to live feed HackRF through a libhackrf python wrapper layer, rather than generating an IQ sample files that would later be hackrf_transfer’d.

HackRF python wrapper

HackRF python wrapper pyhackrf.py is included in this repository but is also proposed to be merged into hackRF main repository: https://github.com/greatscottgadgets/hackrf/pull/1058
If the pull request get accepted, file pyhackrf.py will be removed from this repo.
This repo only uses TX feature of the python wrapper, but RX is also possible (see examples in the PR)

At time of writting this guide, I also believe there is a regression in libhackrf which should be solved by PR: https://github.com/greatscottgadgets/hackrf/pull/1057
This is still under review from greatscottgadgets maintainers but code in this repo is tested with the PR included.
I have not tested it with older/officiel releases of hackrf drivers/dev lib versions.

Software architecture

The workflow is divided between 3 execution threads:

• main thread wich performs all initializations and control user inputs (mainly start / stop simulation for now)
• hackrf broadcasting thread which pump encoded messages and send them over the air with a predefined schedule
• trajectory simulation thread which feed brodcasting thread with encoded messages matching a real time simulated trajectory

The message encoding is splitted into mode S “frame encoding” and “low level encoding” which handles PPM modulation and conversion to hackRF IQ sample format.
Software source code structure tries to reflect those 2 different layers.

So far only “simple” simulated trajectories are available, but one can easily extend/fork behaviour to e.g. have a flight informations coming from a flight simulator (X-plane would be pretty well suited for that purpose through it’s UDP aircraft state broadcast facility) or use actual sensors to feed live data.

Usage and RF broadcast disclaimer

Usage can be demonstrated together with dump1090-mutability or dump1090-fa and associated webservers or text message views.

Repository source code is tuned for a 1090 MHz brodcast with direct wire feed to a receiver SDR dongle (no over the air broadcast).
The hardware setup I’m using is pictured below. Please note the RF attenuators (-20dB and -30dB).
The extra 1090MHz filter is probably not requiered as the flight aware dongle already features 1090 MHz filtering.
My HackRF is fitted with a 0.5 ppm TCXO

The default hackrf settings in repo are