Target Identification Using Modeled Radar Cross Sections and a Coordinated Flight Model

Passive radar is a rapidly emerging technology with many distinct advantages over traditional radar. Its exploitation of “illuminators of opportunity” renders it covert, as well as less expensive. Several passive radar systems, such as Lockheed Martin’s Silent Sentry and John Sahr’s Manastash Ridge Radar at the University of Washington, are already capable of detecting and tracking aircraft. Our goal is to enhance such systems with the addition of automatic target recognition capabilities. We propose conducting target recognition by comparing the radar cross section (RCS) of detected targets to the simulated RCS of known targets. Since RCS is the sole parameter for identification, it is imperative that RCS be modeled accurately. This approach begins by simulating the RCS of aircraft in the target class using the Fast Illinois Solver Code (FISC). These FISC results are then compiled into a database, providing aircraft RCS from a variety of incident and observed angles. Since RCS is highly aspect dependent, aircraft orientation must be known. This challenge is met with a coordinated flight model, which estimates aircraft orientation from aircraft position. To provide further accuracy, the RCS is scaled to account for propagation losses and the receiving antenna gain; these tasks are accomplished using the Advanced Refractive Effects Prediction System (AREPS) and Numerical Electromagnetic Code (NEC2), respectively. The Rician model then compares the simulated RCS of known targets to the RCS arriving at the receiver. Target identification results from this comparison. Thus far, the results are encouraging. The algorithm is able to correctly identify aircraft in the target class with exceptional accuracy at the anticipated noise levels. Performance declines as the noise power surpasses the maximum signal power, but even then, little degradation is noted due to the use of the coordinated flight model.