- Receive ADS-B messages from dump1090 TCP BaseStation output (default port 30003).
- Build up full flight details from multiple messages.
- Track and publish to MQTT within threshold in nautical miles.
adsb2mqtt has multiple settings you can set either direclty in appsettings.json or override with environment variables. I run it on a Docker container and use environment variables. See below for applications.
Setting | Default | Description |
---|---|---|
BEAST_HOST | localhost | dump1090 host |
BEAST_PORT | 30003 | dump1090 TCP BaseStation output port |
MQTT_SERVER | localhost | MQTT host |
MQTT_USE_TLS | false | Connect to MQTT using TLS (mqtts) |
MQTT_PORT | 1883 | MQTT port. |
MQTT_USERNAME | MQTT user name | |
MQTT_PASSWORD | MQTT password | |
TOPIC_BASE | ADSB/flight | MQTT topic containing JSON payload |
LATITUDE | 47.9073 | Set to your latitude |
LONGITUDE | -122.2821 | Set to your longitude |
RADIUS_NM | 3.1 | Nautical mile radial distance threshold to publish flight |
AIRCRAFT_DB_PATH | /usr/share/dump1090-fa/html/db | dump1090-fa flight database |
I've tested both local MQTT and remote MQTTS Mosquitto hosts. The included Dockerfile is for an AMD64 server but I've run this standalone on an RPi3. Link to the .NET install script instructions, including for a Raspbery Pi running Linux.
My main application of adsb2mqtt is a seperate MQTT Python client that uses text-to-speech (TTS) to periodically translate the ADSB/flight payload to a wave file using Microsoft Cognitive Services. Many other TTS options exist. Then publish another MQTT topic with the payload containing the wave filename.
Another option is to generate the wave to a media directory accessible to HA and write an Automation to play the wave. I haven't played with HA's recent improvements for TTS though -- might be able to skip creating the wave file. One other option is a simple mosquitto_sub client on NetBSD to play the wave file. Something roughly like this:
#!/bin/sh
subscribe_speech () {
mosquitto_sub -v -t "ADSB/speech/wave" -u username -P password -h localhost -p 1883 | while read msg
do
wave=$(echo $msg | awk '{print $2}')
curl --silent --output - -X POST -H "Content-Type: text/plain" --data "${wave}" "http://host_serving_tts_wave" | audioplay -d /dev/audio1
done
}
while true
do
subscribe_speech
sleep 5
done
Ultimately, It's a simple pleasure to sit on your patio and hear the departure, airline, altitude (or whatever else you want), of planes you can spot flying over.
I keep RADIUS_NM pretty short, basically I want to see the aircraft enough to roughly make out its type. You can also use an ICAO database to get the plane type. Where I live there's a route that planes fly within a 3nm radius usually under 5000ft MSL before landing at the nearby airport.
You don't need to set LATITUDE and LONGITUDE to your precise location (unlike with say dump1090-fa). Since I usually view planes on my porch looking North, I increase the LATITUDE a bit since my house blocks my view of planes approaching from the South.
NetBSD 10_RC on a Pine64 Rock64 running dump1090-fa
I replaced an aging RPi3 running piaware with a 4gb Rock64 running dump1090-fa. Here's a repo for getting this to run: rtl-sdr-bsd. Here are more details of the overall project: Experimenting with RTL-SDR on NetBSD 10.
FreeBSD 14 on the Rock64 running dump1090-fa
I've since switched from NetBSD 10 to running FreeBSD 14 on the Rock64 device and this resulted in a more stable host for dump1090. It's mainly about better support for RTL-SDR generally with USB on FreeBSD. FreeBSD 14 also now supports .NET 8.
Regardless of your O/S choice, adsb2mqtt should be able to connect to dump1090 and generate MQTT messages.