Beamspace Adaptive Processing for Phased-Array Weather Radars

The Multifunction Phased Array Radar (MPAR) system is required to concurrently perform aircraft and weather surveillance functions with update times of 4.8 seconds and one minute, respectively. To meet these demanding update-time requirements, an MPAR system will likely employ multiple simultaneous beams. One way to achieve this is to transmit a spoiled beam and use digital beamforming to form multiple simultaneous receive beams. However, this solution has difficulties in meeting the data-quality requirements for weather observations due to the increased sidelobe levels in the two-way radiation pattern compared to similarly sized dish antennas. Since weather signals have a large dynamic range, low sidelobe levels are key to prevent volumes with high reflectivity biasing the estimates for volumes with low reflectivity. In the single-beam case, the transmit and receive radiation patterns combine to drive down the two-way sidelobe levels. With a spoiled transmit beam and simultaneous receive beams, the transmit beam does not contribute significantly to lower the two-way sidelobe levels, and an aggressive taper must be used during beamforming to drive down the sidelobe levels of the receive beam such that the two-way sidelobe levels meet requirements. This aggressive taper causes a loss of sensitivity and a loss of spatial resolution (i.e., a broader beamwidth). An alternative solution is to use adaptive beamforming so that the sidelobe levels are automatically lowered in the directions with significant weather returns. However, the adaptive nature of the radiation pattern causes reflectivity calibration to become more difficult compared to the calibration process for a fixed radiation pattern. This paper presents an adaptive beamforming algorithm that operates in beamspace and solves the reflectivity calibration problem while producing accurate estimates of the meteorological variables. Unlike traditional adaptive beamforming algorithms, beamspace processing operates on the received signals from a set of deterministic beams, which allows beamspace processing to produce calibrated meteorological data while retaining the capability of performing spatial filtering of interfering signals. In addition, compared to element-space adaptive beamforming algorithms, beamspace adaptive processing is computationally simpler and requires fewer samples to achieve comparable estimation accuracy, which makes it more attractive for real-time implementation. In this work, simulations are used to quantify the superior performance of the proposed algorithm compared to conventional dish antenna systems and also to traditional Fourier and Capon beamforming on phased-array antenna systems.