Iridium ACARS Position Inference (PoC)

This is a minimal proof-of-concept pipeline that infers aircraft position from Iridium SBD (ACARS) message fragments.

No Doppler.
No multilateration.
No satellite ephemeris.

Position emerges from constraints.

---

What This Does

• Parses Iridium .ul capture data
• Extracts aviation-relevant content (waypoints, routes, etc.)
• Infers a plausible aircraft position
• Outputs Basestation (SBS) format messages
• Streams directly into tools like tar1090

---

Quick Start (10 seconds)

python3 sbd_mo_pipeline_v2_with_basestation.py synthetic_test_data.ul --basestation-out

You should see multiple lines of output like:

MSG,3,1,1,IRDM001,,,,,,,39.5,-82.0,,,,

---

Stream to tar1090

If you have tar1090 / readsb running:

python3 sbd_mo_pipeline_v2_with_basestation.py synthetic_test_data.ul --basestation-out | nc 127.0.0.1 30003

---

Expected Result

• Aircraft appear on the map
• Callsign: IRDM001 or others that match your synthetic_test_data.ul content
• Location: U.S.

This is synthetic data used to validate the pipeline.

---

Files

• 'README.md' - this file
• sbd_mo_pipeline_v2_with_basestation.py — main pipeline
• synthetic_test_data.ul — test input dataset
• 'faa_navaid_database.json' - waypoint/NAVAID database

---

Requirements

• Python 3.x
• nc (netcat) for streaming (optional)
• tar1090 / readsb for visualization (optional)

---

Creating Your Own Input Data

The .ul input used by this PoC is derived from Iridium Toolkit parser output.

In practice, this means:

• Capture Iridium traffic with your preferred front-end
• Run it through the Iridium Toolkit parser
• Use the resulting output.parsed file as the pipeline input

For this proof-of-concept, the .ul file is simply the parsed Iridium Toolkit output renamed for convenience.

Example:

cp output.parsed my_capture.ul
python3 sbd_mo_pipeline_v2_with_basestation.py my_capture.ul --basestation-out

This repository includes synthetic_test_data.ul so you can run the demo immediately without setting up a capture pipeline.

---

Why This Matters

This demonstrates a different approach to geolocation:

Observations are treated as constraints, not answers.

Given enough constraints, position and trajectory emerge—even when direct measurements are absent.

---

Status

• Synthetic dataset: ✅ working
• Real Iridium capture: in progress
• Live streaming: next step

---

Contributing / Interested?

If you find this interesting or want to experiment, feel free to open an issue or reach out.

More to come.