hellmarbecker / plane-kafka

Software define radio plane tracking into KSQL Kafka queries

Plane Kafka - Plane tracking with KSQL and a Raspberry Pi

Raspberry Pi Setup

• If you don't have a Raspberry Pi you can skip to Docker Setup and use the sample data
• Clone this repo on the Raspberry Pi
• Install dump1090
• Set the IP address below to the docker host (and not the IP address of the Raspberry Pi). This is where you'll transmit the messages to Kafka

# On the Raspberry Pi 
cd raspberry-pi
export HOST_IP=192.168.1.129 # Docker host
./plane-kafka.py

Docker Setup

On your host (probably your laptop or PC). Clone this repo

# Start the containers 
docker-compose up -d

Prepare Database Files

You will need a database of icao-to-aircraft mappings (in icao-to-aircraft.json) and a database of callsigns (callsign-details.json). A good source of data is https://openflights.org/data.html where you can find aircraft data suitable for your region

If you are in a hurry, icao-to-aircraft.json.sample and callsign-details.json.sample provide you basic records to experiment

Run the application

docker-compose exec confluent bash

And within the container

confluent start
cd /scripts
./01_setup_topics

If you do not have database files, copy the sample files

cp -i icao-to-aircraft.json.sample icao-to-aircraft.json
cp -i callsign-details.json.sample callsign-details.json

Now load the files. This will load data into the icao-to-aircraft and callsign-details topics

./02_do_load

And then run ksql. If you recieve parse errors, try running the KSQL statments one by one manually instead of the entire script as one.

ksql
-- paste the commands from file 03_ksql.sql
exit

Still within the container finish the remaining setup steps

./04_elastic_dynamic_template
./05_set_connect

Setup kibana

• Navigate to http://localhost:5601
• Create indexes for locationtable and callsigntable. Each should have a EVENT_TS field marked as a timestamp
• Use the Kibana managment page to import saved objects in 06_kibana_export.json