Balanced Correlation Receivers with Applications to Precision Radiometry

This thesis is focused on the development of novel polarimetric techniques and radio receiver designs for precision radiometry applied to 21 cm cosmology and mapping of the radio synchrotron background. The work is organized around two overarching projects, the cosmic twilight polarimeter and the GBT 310 MHz Absolute Mapping project. The astrophysical context for these experiments will be introduced in Chapter 1, where I present a brief history of the universe, including the formation of the first stars and galaxies, the nature of the low-frequency radio sky, and the basics of radio receivers.

In Chapter 2, I describe experimental work on the dynamically induced polarization effect initially proposed by my predecessor, Bang Nhan. This polarimetric approach is intended to reduce the beam-weighted foreground degeneracy typical in global 21 cm experiments by using the polarization leakage of the antenna to provide extra information on the anisotropic component of the sky. I provide a mathematical description of the antenna polarization leakage and associated dynamically induced polarization using the Mueller matrix representation, which intuitively shows how the instrument responds to arbitrarily polarized light from areas of the sky. I created a series of observation simulations using antenna beam models and the all-sky Haslam map, including atmospheric propagation effects and physical perturbations to the system. These simulations were then compared with experimental data provided by a repurposed PAPER antenna and receiver system that I deployed at Green Bank Observatory from fall 2019 to spring 2020. These observations qualitatively agree with simulations, but large deviations exist that may suggest missing contributions from sky polarization, propagation or instrumental effects.

In Chapter 3, I present the theory of operation for a novel type of radiometer, a Balanced Correlation Receiver (BCR), that virtually eliminates all measurement offset biases and gain instability (1/f noise).
The BCR accomplishes this through a unique digital synthesis method for forming the antenna beam directly from cross-correlations between the balanced outputs of the feed, thereby avoiding any directly shared circuitry between the feed points of the antenna and the receiver (e.g., baluns, transmission lines, power splitters and hybrids) that could introduce common-mode noise. I derive how different dipole beams can be formed from combinations of auto and cross-correlations of redundant-in-polarization e-field probes (e.g., the two monopole arms of a dipole). Additionally, I extend this analysis to the cross-dipole and apply digital polarization synthesis methods to derive the full-stokes polarization response and sensitivity for a BCR and limitations thereof. This approach is essential for the planned 310 MHz Sky Map and is the theoretical basis of the GBT310 instrument in Chapter 4.

In Chapter 4, I motivate the need for a new low-frequency sky map with absolute, zero-level calibration as the primary goal and which requires the Green Bank Telescope and a custom-built receiver. I present the scientific utility of such a new map, planned mapping strategy for the initially awarded NSF grant (PI -- Jack Singal) and observational requirements for the custom instrument, the GBT 310 MHz receiver that employs the balanced correlation architecture from Chapter 3. The instrument development for the GBT310 includes the design, construction, lab testing, calibration and field commissioning of all critical sub-systems: the feed, frontend receiver module, data acquisition, monitoring and control, and system software. Finally, I present current results from the ground commission and testing of the GBT310 in preparation for first-light on the GBT.

Finally, Chapter 6 provides a summary of the key results and related future work. Additionally, the appendices contain detailed information about the design and specification of GBT310 components.