Statistical Analysis and Flight Route Extraction from Automatic Dependent Surveillance-Broadcast Data

Modern air traffic management considers wide integration of Automatic Dependent Surveillance-Broadcast (ADS-B) technology in existing air navigation system. Each airspace user should be equipped with a transponder of mode 1090ES for sharing position reports with all other airspace users. Air navigation service providers support ground networks of software defined radios for collecting position reports via 1090ES communication channel. Statistical analysis of historical ADS-B data can be useful for delays prediction and safety level estimation.In this paper, we propose to use ADS-B data for statistical analysis of time of arrival and delays at particular waypoint of airplane flight route. The study includes waypoints identification and statistical analysis of delays for a set of waypoints. All flights within controlled airspace should be performed within the network of air traffic service (ATS) routes, except free-routes airspace. Thus, it makes sense to associate surveillance data with a network of ATS routes to identify cleared route. Air traffic usually deviates from the flight plan center line within the perils of vertical and lateral separation minimums. In the paper, we propose a sequential algorithm for waypoints identification by minimum distance from the network of ATS routes. Also, a network of ATS routes is represented in the form of an undirected graph with nodes associated with waypoints and edges with linear airway between waypoints.Obtained set of waypoints is used in the statistical analysis of time of arrival with a kernel density estimator for each waypoint separately. Distribution of kernel density functions for time of arrival at each waypoint can be useful for total delays prediction and analysis of causes acted at each edge of the extracted flight route.Calculated airplane deviations from ATS routes are used for risks of separation loss estimation. A Triple Univariate Generalized Error Distribution function is used in the analysis of lateral deviations.In the numerical demonstration, we use archived ADS-B data of Ukrainian local flight connection between Boryspil and L’viv airports for delays and safety analysis.