Using Autoencoders To Learn Interesting Features For Detecting Surveillance Aircraft

TL;DR

This paper presents an LSTM-based sequence autoencoder approach to detect surveillance aircraft by learning features from ADS-B flight data, leveraging LSTM's ability to model variable-length, irregular, and dependent time series.

Contribution

It introduces a novel application of LSTM autoencoders for surveillance aircraft detection using ADS-B data, addressing challenges of variable length and irregular sampling.

Findings

LSTM autoencoders effectively learn features for aircraft detection.

The method handles variable-length and irregular time series data.

Potential for improved surveillance detection accuracy.

Abstract

This paper explores using a Long short-term memory (LSTM) based sequence autoencoder to learn interesting features for detecting surveillance aircraft using ADS-B flight data. An aircraft periodically broadcasts ADS-B (Automatic Dependent Surveillance - Broadcast) data to ground receivers. The ability of LSTM networks to model varying length time series data and remember dependencies that span across events makes it an ideal candidate for implementing a sequence autoencoder for ADS-B data because of its possible variable length time series, irregular sampling and dependencies that span across events.