Relaxed concentrated MLE for robust calibration of radio interferometers

TL;DR

This paper introduces a robust calibration method for radio interferometers using a relaxed concentrated MLE approach that accounts for perturbations and SIRP noise, providing closed-form solutions for key parameters.

Contribution

It proposes a novel iterative relaxed concentrated MLE algorithm with closed-form expressions, improving calibration robustness in radio interferometry.

Findings

Effective calibration under perturbations

Closed-form solutions for unknown parameters

Enhanced robustness to noise

Abstract

In this paper, we investigate the calibration of radio interferometers in which Jones matrices are considered to model the interaction between the incident electromagnetic field and the antennas of each station. Specifically, perturbation effects are introduced along the signal path, leading to the conversion of the plane wave into an electric voltage by the receptor. In order to design a robust estimator, the noise is assumed to follow a spherically invariant random process (SIRP). The derived algorithm is based on an iterative relaxed concentrated maximum likelihood estimator (MLE), for which closed-form expressions are obtained for most of the unknown parameters.