Calibration methods in large interferometric radiometers devoted to earth observation

Abstract

This paper starts with a summary and classification of errors of any kind that corrupt the fundamental measurement performed by an InR (interferometric radiometer), the so-called visibilities. This summary is based on prior works from the authors concerning end-to-end rnodeling of the instrument. The paper follows with a trade-off analysis of the capability of different calibration approaches to remove those errors and, at the end, to recover a map of brightness temperatures. In order to perform the trade-off analysis, extensive simulations have been undertaken to analyze the approaches found in the literature: the redundant space method, used in radio astronomy, the G-matrix method, used in ESTAR, and the Noise injection method. both centralized and distributed, proposed for MIRAS. A short description is made of each one and a comparative table is presented. The trade off is presented in terms of what kind of errors can be removed by each method, hardware requirements, robustness and on-ground/on-board input data required to perform such error correction.

Peer Reviewed

Postprint (published version)