Data Analysis in Global 21-cm Experiments: Chromatic Effects and Signal Reconstruction

This thesis investigates chromatic effects that can distort the global 21-cm signal and introduces a flexible signal model designed to capture potential features arising from unexpected physics. The work is carried out within the framework of the REACH experiment and its data analysis pipeline. Part I includes two chapters. Chapter 1 provides an introduction to 21-cm cosmology, and outlines both past and upcoming experimental efforts. Chapter 2 presents the numerical and statistical methods adopted throughout this thesis. Part II and Part III contain the original research contributions, each chapter beginning with a contextual introduction. Part II focuses on chromatic effects. Chapter 3, published as Shen et al. (2022), addresses the modelling of time-varying ionospheric effects and demonstrates that these effects do not average out over time, highlighting the importance of accounting for temporal variation in data analysis. Chapter 4, published as Shen et al. (2024), examines contamination from Galactic linear polarisation. Using three distinct models, the REACH pipeline is stress-tested and shown to recover injected global 21-cm signals in most cases, despite unaccounted-for polarisation. In Part III, Chapter 5, the FlexKnot signal model is presented. This model is characterised by a twice-integrated cubic function built from freely positioned knots interpolated by piecewise linear splines. It is shown to effectively recover absorption features that may originate from unaccounted-for physics in the global 21-cm signal, performing robustly within the REACH data analysis framework. This work is published as Shen et al. (2025). Finally, Part IV summarises the main conclusions and outlines potential directions for furthering these efforts.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2025

Author dc:creator

• Shen, Emma

Advisors dc:contributor.advisor

• De Lera Acedo, Eloy
• Fialkov, Anastasia

• Cosmology
• Bayesian statistics
• Data Analysis

Licence

• https://www.repository.cam.ac.uk/bitstreams/496855ec-e659-4522-af4b-9d8a87c52402/download
• http://purl.org/NET/rdflicense/allrightsreserved

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.125277

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/395929