Dimension embedding for big data in radio interferometry

As radio telescope data is now increasingly high- dimensional, data reduction has become essential. This abstract describes dimension embedding techniques that are being explored as a way to reduce data size and, consequently, computational load while preserving accurate signal reconstruction (in terms of SNR and dynamic range) in the context of radio interferometric imaging. In this work, dimension embeddings are being designed and evaluated for their ability to preserve information after embedding. Preliminary results suggest that random Gaussian embeddings may provide significant improvement with respect to standard gridding.