Efficient correction for both direction-dependent and baseline-dependent effects in interferometric imaging: An A-stacking framework

Artikel i vetenskaplig tidskrift, 2015

A general framework is presented for modeling direction-dependent effects that are also baseline-dependent, as part of the calibration and imaging process. Within this framework such effects are represented as a parametric linear model in which basis functions account for direction dependence, whereas expansion coefficients account for the baseline dependence. This separation enables the use of a multiple fast Fourier transform-based implementation of the forward calculation (sky to visibility) in a manner similar to the W-stacking solution for non-coplanar baselines, and offers a potential improvement in computational efficiency in scenarios where the gridding operation in a convolution-based approach to direction-dependent effects may be too costly. Two novel imaging approaches that are possible within this framework are also presented.