Estimation of HII Bubble Size Distribution from 21cm Power Spectrum with Artificial Neural Networks
Abstract
The bubble size distribution of ionized hydrogen regions probes the information about the morphology of \ bubbles during the reionization. Conventionally, the \ bubble size distribution can be derived from the tomographic imaging data of the redshifted 21~cm signal from the epoch of reionization, which, however, is observationally challenging even for the upcoming large radio interferometer arrays. Given that these interferometers promise to measure the 21~cm power spectrum accurately, we propose a new method, which is based on the artificial neural networks (ANN), to reconstruct the \ bubble size distribution from the 21~cm power spectrum. We demonstrate that the reconstruction from the 21~cm power spectrum can be almost as accurate as directly measured from the imaging data with the fractional error 10\%, even with thermal noise at the sensitivity level of the Square Kilometre Array. Nevertheless, the reconstruction implicitly exploits the modelling in reionization simulations, and hence the recovered \ bubble size distribution is not an independent summary statistic from the power spectrum, and should be used only as the indicator for understanding \ bubble morphology and its evolution.