7–11 Jul 2025
Teaching and Learning Centre (TLC)
Europe/London timezone

Bayesian Power Spectrum estimation with built-in systematics modelling for the HERA array

Not scheduled
1h 30m
Teaching and Learning Centre (TLC)

Teaching and Learning Centre (TLC)

Durham University South Road Durham DH1 3LS
Poster Radio Astronomy in the build up to the SKAO Radio Astronomy in the build up to the SKAO

Speaker

Sohini Dutta (The University of Manchester)

Description

The redshifted 21-cm signal from Cosmic Dawn (CD) and the Epoch of Reionization (EoR) is challenging to observe due to a variety of systematic effects that radio interferometers must contend with. The modulated foreground, which is notoriously difficult to model; substantial gaps caused by the excision of regions necessary due to radio frequency interference (RFI); and artifacts resulting from instrumental systematics all compromise the observable of interest. In this work, we present a novel method of recovering data buried under instrumental systematics, primarily cable reflections and mutual coupling. We use a Bayesian statistical method that couples the techniques of Gaussian constrained realizations (GCR) and Gibbs sampling to overcome uncertain foreground models, gappy data after RFI excision, and to subsequently estimate the EoR delay power-spectrum. We demonstrate the application of this technique on simulated data that mimic the Hydrogen Epoch of Reionization Array (HERA).

Primary author

Sohini Dutta (The University of Manchester)

Co-authors

Dr Jacob Burba Katrine Glasscock (University of Manchester) Dr Michael Wilensky Dr Phil Bull Dr Zheng Zhang

Presentation materials

There are no materials yet.