Speaker
Description
The Epoch of Reionization (EoR) and Cosmic Dawn (CD) play crucial roles in shaping the early Universe during its initial billion years. Despite their significance, the characteristics of the Intergalactic Medium (IGM) during these epochs remain uncertain and require further observational validation. Current and upcoming radio telescopes, such as EDGES, SARAS, MWA, and SKA, aim to detect the redshifted HI 21-cm signal, a key objective in understanding these cosmic periods. However, challenges arise from systematic errors and the need for precise foreground removal.
Especially at lower frequencies, the Earth's ionosphere introduces significant distortions, creating direction-dependent effects on the signal. Therefore, it is imperative to comprehend the impact of each source of corruption when employing non-parametric techniques for accurate 21cm signal detection. This presentation is divided into two parts:
i) We will introduce an end-to-end pipeline developed to investigate the influence of bright foreground, primary beam, ionosphere, and other systematics on extracting the redshifted HI 21cm signal from upcoming interferometric experiments like SKA-1 Low. This pipeline incorporates various signal extraction methods, with a specific focus on power-spectrum estimation. Simulation is employed to understand and quantify how different ionospheric scale structures affect the precision of 21-cm signal power spectrum estimation in synthetic observations with SKA1-Low.
ii) We will discuss parameter estimation using Artificial Neural Networks (ANN) and Markov Chain Monte Carlo (MCMC) to extract the HI 21-cm power spectrum and associated parameters from observed sky signals using this end to end pipeline. This involves considering the HI signal, actual foregrounds, calibration and position errors, as well as systematic effects.
The developed end-to-end pipeline enhances the accuracy of ground-based data analysis for upcoming radio interferometric studies like SKA, MWA, and HERA. This improvement contributes to a better understanding of ionospheric effects and systematic error tolerance levels across different arrays and their impact on parameter estimation.
