OCREN: A countermeasure for ADS-B attacks in the context of air traffic control

Automatic dependent surveillance–broadcast (ADS-B) is a surveillance technology widely endorsed by the International Civil Aviation Organization (ICAO). It has seen extensive use in both commercial and general aviation, playing a vital role in supporting air traffic control (ATC) operations. However, the inherent openness of the ADS-B protocol leaves it vulnerable to malicious network attacks. Previous research has largely overlooked ATC application scenarios in commercial air transportation, and existing detection algorithms suffer from low attack detection rates. Furthermore, current research lacks effective and reliable methods for responding to identified attacks. This paper primarily investigates ADS-B attack detection and recovery within the context of ATC. We integrate flight plans with ADS-B data to create a flight data restriction domain and employ the online classification restriction-extreme learning machine (OCR-ELM) model to detect and flag abnormal data streams. Subsequently, the nonlinear autoregressive neural network (NARX) model, combined with a recovery strategy, is utilized to restore the flagged abnormal data. The proposed method enables precise and rapid identification of abnormal targets in ADS-B data and eases targeted recovery of the affected data points. We employed real flight data to simulate anomalous data caused by various malicious attacks, evaluating detection and recovery performance using several performance metrics. The results show that our method achieves an average detection accuracy of 98.6%, a false positive rate of 1.69%, a false negative rate of 0.93%, and an average structural similarity index of 0.9895 for the recovered data. This method ensures the security of information and continuous accessibility of ADS-B, thus enhancing the operational safety and robustness of air traffic control.