Radar Data Tracking Using Minimum Spanning Tree-Based Clustering Algorithm

This paper discusses a novel approach to associate and refine aircraft track data from multiple radar sites. The approach provides enhanced aircraft track accuracy and time synchronization that is compatible with modern air traffic management analysis and sim-ulation tools. Unlike existing approaches where the number of aircraft in the radar data must be assumed, this approach requires no such prior knowledge. While commercial air-craft provide ID tags captured in the radar data in the form of Mode 3 transponder codes, general aviation often lacks such transponders, which precludes using the number of codes sensed to count the number of aircraft in the data. To meet this challenge, an approach to track an unknown number of unidentified aircraft using a clustering algorithm is proposed. The paper presents a method to relate aircraft between consecutive time frames and refine the trajectories of those vehicles. Experimental results from evaluating the algorithm and demonstrating its viability are provided. Nomenclature σL Standard deviation for edge lengths [nmi] Θh Maximum allowable difference between headings of two aircraft [ Θp Maximum allowable distance between the real position and the expected position of an aircraft [nmi] di Length of the ith edge in a spanning tree [nmi] Lmax Maximum distance in a cluster [nmi] mL Average of edge lengths [nmi] I