A New Association Approach for Multi-Sensor Air Traffic Surveillance Data Based on Deep Neural Networks
This repository contains the implementation of the M-SIOTA algorithm as described in the paper “A New Association Approach for Multi-Sensor Air Traffic Surveillance Data Based on Deep Neural Networks”.
The required packages for training and running the algorithm are in the requirements.txt file.
To run the association algorithm, use the Jupiter notebook main_associator.ipynb. The notebook expects a .csv file in the opensky directory. The route and filename of the dataset can be changed in the default_config.json file. The 24h State Vector from OpenSky Network file format is used.