Atmospheric Radar Research Center

Phased Array Radar operates at 3.2 GHz and utilizes the same Klystron transmitter used on the WSR-88D radar network and thus shares many of the same characteristics. PAR employs the SPY-1A phased array antenna used on Aegis-class cruisers and destroyers and can perform a sector scan of 90° in both azimuth and elevation through electronic scanning. Its 4,352 elements produce a beamwidth of 1.5° at broadside and 2.1° at the maximum o-broadside pointing angle of 45°. The antenna is mounted on a pedestal with rotational speed of up to 18° per second, which allows the radar to complete a full 360° volume coverage in less than 60 sec.

Building on previous work of the investigators [Yu et al., 2000; Cheong et al., 2004] and with the goal of having a controlled experimental data source, a sophisticated time-series radar simulator will be created to ingest high-resolution, three-dimensional, meteorological fields and to generate synthetic radar time-series data.

The concept is based on emulating volume scattering from the atmosphere using numerous (>10,000) point scatterers. The characteristics (motion, reflectivity, etc.) of these point scatterers are determined by the input meteorological fields. By coherently summing the electromagnetic signals backscattered from each of the point targets, it is possible to generate the time-series signal for any point on the receive antenna, which could consist of an array of elements, for example. Therefore, it is possible to generate independent baseband signals for each element of a phased array, allowing studies of digital beamforming, array configurations, etc.
The simulator will be designed with the flexibility to match the characteristics of PAR, MPAR, or any other more-conventional radar system, such as the WSR-88D. Beamwidth, operating frequency, range weighting, and antenna pattern, will all be controlled at the input stage of the simulator. In order to support some of the proposed MPAR applications, frequency chirp and digital beamforming capabilities will be made possible.

They will be obtained from the high-resolution (25 m and 1 sec) ARPS (Advanced Regional Prediction System) numerical simulation model developed at the Center for the Analysis and Prediction of Storms (CAPS) at OU [Xue et al., 2000, 2001, 2003]. A conceptual diagram of the time-series simulator is shown in Figure 2.

Although other time-series simulators have been devised [Capsoni and D’Amico, 1998; Capsoni et al., 2001], this would be the first effort to use realistic storm-scale meteorological fields for the governing structure and dynamics. Furthermore, it would be the first simulation of a phased array weather radar. In addition to atmospheric scattering, it is important to have the capability to introduce ground and airborne clutter targets to the simulation.

Since the simulator is based on point targets, it is relatively simple to introduce several highly reflective targets with independent motion characteristics, facilitating the test of various multi-function algorithms. These simulated signals will have arbitrary maneuver characteristics, allowing study of both cooperative and non-cooperative target tracking algorithms in a weather-cluttered environment. Finally, ground clutter targets can be placed in the radar field of view facilitating the study of not only refractivity retrieval but also the development of clutter mitigation techniques. With this flexibility, it is anticipated, and in fact planned, that the time-series simulator will be exploited for several phases of the proposed work where actual data are non-existent.