Building a Meshtastic Network Monitoring Suite

Context

Meshtastic is an open-source mesh radio platform used for off-grid communication. The existing Meshtastic MQTT Explorer visualizes live node data, but it does not answer two critical planning questions: where does each node actually cover? and where are the gaps if a node goes down?

I built meshnetwork as a complete monitoring extension — radio coverage prediction, redundancy scoring, and a dedicated map viewer — all containerized and running alongside the upstream explorer with zero modifications to its codebase.

Timeline: Personal project, 2025
Repo: github.com/soufian-elouazzani/meshnetwork · Docs: meshnetwork-docs

What I built

• Coverage Service — Python daemon that uses SPLAT! and NASA SRTM terrain data to predict radio coverage per node and export GeoJSON polygons (~15 seconds per node with terrain caching).
• Redundancy Service — Detects overlaps between coverage zones, scores redundancy on a 1–5 scale, and highlights weak points where a single node failure would leave a gap.
• Vue.js microfrontend — Leaflet-based real-time map viewer on port 3001, separate from the main Blazor front end on port 80.
• Docker Compose stack — PostgreSQL, Mosquitto MQTT, Python daemons, .NET 9 explorer, and Vue viewer orchestrated together.

Technical decisions

• Extend, don't fork — Integrating via MQTT topics and shared PostgreSQL rather than patching upstream keeps updates painless and demonstrates microservice boundaries.
• SPLAT! + SRTM for terrain-aware coverage — Flat-earth approximations fail in mountainous regions; real elevation data produces polygons operators can trust.
• Separate Vue viewer — The redundancy and coverage layers need different interaction patterns than the live node explorer; a dedicated front end keeps each UI focused.

Results and takeaways

• Operators can plan node placement before deploying hardware in the field.
• Redundancy scoring turns subjective "we probably have overlap" into actionable weak-point maps.
• For clients and recruiters: this shows end-to-end system design — backend services, data pipelines, container orchestration, and a polished map UI.

Stack

Quick start

git clone https://github.com/soufian-elouazzani/meshnetwork.git
cd meshnetwork/docker
sudo docker compose up -d

• Main map: http://localhost
• Coverage viewer: http://localhost:3001

See the full README and architecture docs for diagrams and screenshots.