%0 Generic %1 shimabukuro2020beyond %A Shimabukuro, Hayato %A Mao, Yi %A Tan, Jianrong %D 2020 %K library %T Beyond power spectrum I: recovering HII bubble size distribution from 21cm power spectrum with artificial neural network %U http://arxiv.org/abs/2002.08238 %X The bubble size distribution of ionized hydrogen regions, which could be derived from the tomographic imaging data of the redshifted 21~cm signal from the epoch of reionization, probes the information about the morphology of \HII\ bubbles during the reionization. However, 21~cm imaging is observationally very challenging even for the upcoming large radio interferometers. 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 \HII\ 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, systematic errors might arise from approximations made in reionization simulations used for training the ANN. This paper, as the first in a series, exemplifies the possibility of recovering from the 21~cm power spectrum with ANN additional statistics of cosmic reionization that could not otherwise be inferred from the power spectrum analysis directly in the conventional methods.

@misc{shimabukuro2020beyond, abstract = {The bubble size distribution of ionized hydrogen regions, which could be derived from the tomographic imaging data of the redshifted 21~cm signal from the epoch of reionization, probes the information about the morphology of \HII\ bubbles during the reionization. However, 21~cm imaging is observationally very challenging even for the upcoming large radio interferometers. 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 \HII\ 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 $\lesssim 10\%$, even with thermal noise at the sensitivity level of the Square Kilometre Array. Nevertheless, systematic errors might arise from approximations made in reionization simulations used for training the ANN. This paper, as the first in a series, exemplifies the possibility of recovering from the 21~cm power spectrum with ANN additional statistics of cosmic reionization that could not otherwise be inferred from the power spectrum analysis directly in the conventional methods.}, added-at = {2020-02-20T05:02:15.000+0100}, author = {Shimabukuro, Hayato and Mao, Yi and Tan, Jianrong}, biburl = {https://www.bibsonomy.org/bibtex/2128dd78fc3e9c0d8c6987be6ddeac5fa/gpkulkarni}, description = {Beyond power spectrum I: recovering HII bubble size distribution from 21cm power spectrum with artificial neural network}, interhash = {79fcccd6b4e078fa40dabf3d694f5101}, intrahash = {128dd78fc3e9c0d8c6987be6ddeac5fa}, keywords = {library}, note = {cite arxiv:2002.08238Comment: 12 pages, 12 figures. Submitted to MNRAS}, timestamp = {2020-02-20T05:02:15.000+0100}, title = {Beyond power spectrum I: recovering HII bubble size distribution from 21cm power spectrum with artificial neural network}, url = {http://arxiv.org/abs/2002.08238}, year = 2020 }