// the find
IronGiu/MeshStation
Meshtastic SDR Analyzer & Desktop GUI. A tool to analyze the meshtastic network graphically with a simple sdr.
MeshStation is a desktop GUI that decodes live Meshtastic LoRa traffic straight off the air using an RTL-SDR, no Meshtastic radio required. It's aimed at people doing coverage analysis, network mapping, or just nosing around at what's active on mesh frequencies in their area, via GNU Radio flowgraphs feeding a TCP server into the GUI.
The decoding path is real signal processing, not a wrapper around the official Meshtastic API - the GNU Radio flowgraph (rx_lora_base_engine.py and friends) demodulates LoRa packets directly, which is the actual hard part of this project. It ships locked conda environments and build scripts for five separate platform targets (Windows, Linux x64/ARM64, macOS Intel/Apple Silicon), which is more build engineering than most hobby SDR tools bother with. The engine/GUI split over a TCP server keeps the signal pipeline decoupled from the desktop app, so the two can be iterated on independently. Scanning every modem preset and frequency at once for network discovery is a capability the official Meshtastic client doesn't give you.
There's no test suite anywhere in the tree - for a packet-decoding pipeline that's a real gap, since a misparsed field fails quietly instead of loudly. There's no CI either (the only .github content is FUNDING.yml), so none of those five platform builds are verified automatically and any of them can bit-rot without anyone noticing. The GUI itself is a single top-level MeshStation.py, in contrast to the engine's modular flowgraph structure - the two halves don't show the same level of care. TX support, the feature that would make this more than a passive listener, is explicitly gated behind future funding rather than built, so despite the roadmap language this is receive-only today. Getting it running means picking the right per-OS conda lock file and running a separate builder script per platform, with the actual instructions living in an external wiki instead of the README.