Using Autoencoders To Learn Interesting Features For Detecting\n Surveillance Aircraft

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