Incorporating Aircraft Kinematics and Radar Cross Section into the Performance Prediction of Air Surveillance

The evolution of modern radar is heading toward a networked, multifunctional, adaptive, and cognitive system. The network of software-controllable fast-adapting radars follows a highly complex control and operation logic. It is not straightforward to assess its instantaneous capability to detect, track, and recognize targets. To be able to predict or optimize the system performance, one has to understand its behavior not only on a general level, but also in various operating conditions and considering the target behavior and properties accurately. In this paper, we propose the fusion of radar and tracker recordings with an extensive database of cooperative aircraft navigation recordings and radar cross section data to assess and learn the performance measures for the air surveillance. The main contribution of this paper is the incorporation of the aircraft kinematics, orientation, and radar cross section into an automated measurement-based analysis. We consider the employment of the measurement-based metrics and machine learning in the performance prediction. Simulations and experiments with real-life data demonstrate the feasibility and potential of the proposed concept.