Estimating radar azimuth jitter through analysis of targets of opportunity data

Accurate models of radar errors are important to the safety analysis of air traffic control systems. We detail in this paper a method for estimating radar azimuth errors from raw time-stamped range/azimuth radar reports of aircraft flying through the airspace. Unlike many previous radar error analyses, the method does not require that the true position of the aircraft be known and is therefore not reliant on expensive flight tests. This provides for the ability to analyze large quantities of data at relatively low cost. Our estimations are based on the assumption of straight and level flight at a constant ground speed, the most common flight regime in the en route airspace. The method filters radar data so it contains only aircraft flying straight and level at constant ground speed, and then uses this knowledge of the aircraft's behavior to estimate the true aircraft position and the radar error. The filtering algorithms and estimation method are validated using real radar data, simulated radar targets, and flight test data with GPS truth.