http://arxiv.org/abs/2004.09206
The radio observation of 21cm-line signal from the Epoch of Reionization (EoR) enables us to explore the evolution of galaxies and intergalactic medium in the early universe. However, the detection and imaging of 21cm-line signal are tough due to the foreground and instrumental systematics. In order to overcome these obstacles, as a new approach, we propose to take a cross correlation between observed 21cm-line data and 21cm-line images generated from the distribution of the Lyman-$\alpha$ emitters (LAEs) through machine learning. In order to create 21cm-line maps from LAE distribution, we apply conditional Generative Adversarial Network (cGAN) trained with the results of our numerical simulations. We find that 21cm-line brightness temperature maps and the neutral fraction maps can be well reproduced at large scales. Furthermore, we show that the cross correlation is detectable at $k < 0.2~{\rm Mpc}^{-1}$ by combing 400 hours of MWA Phase II observation and LAE deep survey of the Subaru Hyper Suprime Camera. Our new approach of cross correlation with image construction using the cGAN can not only boost the detectability of EoR 21cm-line signal but also allow us to estimate the 21cm-line auto-power spectrum.
S. Yoshiura, H. Shimabukuro, K. Hasegawa, et. al.
Tue, 21 Apr 20
65/90
Comments: 15 pages, 11 figures, 3 tables, submitted to MNRAS