CAE: Contextual auto-encoder for multivariate time-series anomaly detection in air transportation

The Automatic Dependent Surveillance Broadcast protocol is one of the latest compulsory advances in air surveillance. While it supports tracking ever-growing number aircraft air, also introduces cybersecurity issues that must be mitigated e.g., false data injection attacks where an attacker emits fake surveillance information. recent sources and tools available to obtain flight records allow researchers create datasets develop Machine Learning models capable detecting such anomalies En-Route trajectories. In this context, we propose a novel multivariate anomaly detection model called Discriminatory Auto-Encoder (DAE). It uses baseline regular LSTM-based auto-encoder but with several decoders, each getting specific phase (e.g. climbing, cruising or descending) during its training.To illustrate DAE's efficiency, evaluation dataset was created using real-life as well realistically crafted ones, which DAE three from literature were evaluated. Results show achieves better results both accuracy speed detection. dataset, implementations are online repository, thereby enabling replicability facilitating future experiments.