Improving the Calibration Efficiency of an Array Fed Reflector Antenna through Con-strained Beamforming

Calibrating for the radiation pattern of a multi-beam Phased array feed (PAF) based radio telescope largely depends on the accuracy of the pattern model, and the availability of suitable reference sources to solve for the unknown parameters in the pattern model. It is shown how the efficiency of this pattern calibration for PAF antennas can be improved by conforming the beamformed far field patterns to a two-parameter physics-based analytic reference model through the use of a linearly constrained minimum variance (LCMV) beamformer. Through this approach, which requires only a few calibration measurements, an accurate and simple pattern model is obtained. The effects of the model parameters on the directivity and sidelobe levels of multiple scanned beams are investigated, and these results are used in an example PAF beamformer design for the proposed MeerKAT antenna. Compared to a typically used maximum directivity (MaxDir) beamformer, the proposed constrained beamforming method is able to produce beam patterns over a wide Field-of-View (FoV) that are modeled with a higher degree of accuracy and result in a significant reduction in pattern calibration complexity