Bayesian GPR recovers the 21cm signal within 2σ credible intervals for most k-modes (0.06 to 1.0 h/Mpc) in SKA1-Low simulations that include realistic residual foregrounds and systematics.
Calibration Requirements for Detecting the 21 cm Epoch of Reionization Power Spectrum and Implications for the SKA
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
abstract
21 cm Epoch of Reionization observations promise to transform our understanding of galaxy formation, but these observations are impossible without unprecedented levels of instrument calibration. We present end-to-end simulations of a full EoR power spectrum analysis including all of the major components of a real data processing pipeline: models of astrophysical foregrounds and EoR signal, frequency-dependent instrument effects, sky-based antenna calibration, and the full PS analysis. This study reveals that traditional sky-based per-frequency antenna calibration can only be implemented in EoR measurement analyses if the calibration model is unrealistically accurate. For reasonable levels of catalog completeness, the calibration introduces contamination in otherwise foreground-free power spectrum modes, precluding a PS measurement. We explore the origin of this contamination and potential mitigation techniques. We show that there is a strong joint constraint on the precision of the calibration catalog and the inherent spectral smoothness of antennae, and that this has significant implications for the instrumental design of the SKA and other future EoR observatories.
citation-role summary
background 1
citation-polarity summary
fields
astro-ph.CO 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Simulations show hybrid foreground mitigation (GPR + PCA combined with avoidance) recovers the HI 21cm signal within 2σ for gain calibration errors ≤1% in SKA1-Low AA* observations over 0.05-0.5 Mpc^{-1} scales.
citing papers explorer
-
Mitigating residual foregrounds and systematic errors in SKA1-Low AA* EoR observations via Bayesian Gaussian Process Regression
Bayesian GPR recovers the 21cm signal within 2σ credible intervals for most k-modes (0.06 to 1.0 h/Mpc) in SKA1-Low simulations that include realistic residual foregrounds and systematics.
-
Mitigating gain calibration errors from EoR observations with SKA1-Low AA*
Simulations show hybrid foreground mitigation (GPR + PCA combined with avoidance) recovers the HI 21cm signal within 2σ for gain calibration errors ≤1% in SKA1-Low AA* observations over 0.05-0.5 Mpc^{-1} scales.