The 21 cm signal from the Epoch of Reionization should be observed within the next decade. While a simple statistical detection is expected with SKA pathfinders, the SKA will hopefully produce a full 3D mapping of the signal. To extract from the observed data constraints on the parameters describing the underlying astrophysical processes, inversion methods must be developed. For example, the Markov Chain Monte Carlo method has been successfully applied. Here we test another possible inversion method: artificial neural networks (ANN). We produce a training set which consists of 70 individual sample. Each sample is made of the 21 cm power spectrum at different redshifts produced with the 21cmFast code plus the value of three parameters used in the semi-numerical simulations that describe astrophysical processes. Using this set we train the network to minimize the error between the parameter values it produces as an output and the true values. We explore the impact of the architecture of the network on the quality of the training. Then we test the trained network on the new set of 54 test samples with different values of the parameters. We find that the quality of the parameter reconstruction depends on the sensitivity of the power spectrum to the different parameters at a given redshift, that including thermal noise and sample variance decreases the quality of the reconstruction and that using the power spectrum at several redshifts as an input to the ANN improves the quality of the reconstruction. We conclude that ANNs are a viable inversion method whose main strength is that they require a sparse exploration of the parameter space and thus should be usable with full numerical simulations.

The 21cm line brightness temperature brings rich information about Epoch of Reionizaton (EoR) and high-z universe (Cosmic Dawn and Dark Age). While the power spectrum is a useful tool to investigate the EoR signal statistically, higher-order statistics such as bispectrum are also valuable because the EoR signal is expected to be highly non-Gaussian. In this paper, we develop a formalism to calculate the bispectrum contributed from the thermal noise taking array configularion of telescopes into account, by extending a formalism for the power spectrum. We apply our formalism to the ongoing and future telescopes such as expanded Murchison Widefield Array (MWA), LOw Frequency ARray (LOFAR) , Hydrogen Epoch of Reionization Array (HERA) and Square Kilometre Array (SKA). We find that expanded MWA does not have enough sensitivity to detect the bispectrum signal. On the other hand, LOFAR has better sensitivity and will be able to detect the peaks of the bispectrum as a function of redshift at large scales with comoving wavenumber k 0.03 Mpc −1 . The SKA has enough sensitivity to detect the bispectrum for much smaller scales k 0.3 Mpc −1 and redshift z 20

Redshifted 21cm signal is a promising tool to investigate the state of intergalactic medium (IGM) in the Cosmic Dawn (CD) and Epoch of Reionization(EoR). In our previous work (Shimabukuro et al. 2015), we studied the variance and skewness of the 21cm fluctuations to give a clear interpretation of the 21cm power spectrum and found that skewness is a good indicator of the epoch when X-ray heating becomes effective. Thus, the non-Gaussian feature of the spatial distribution of the 21cm signal is expected to be useful to investigate the astrophysical effects in the CD and EoR. In this paper, in order to investigate such a non-Gaussian feature in more detail, we focus on the bispectrum of the 21cm signal. It is expected that the 21cm brightness temperature bispectrum is produced by non-gaussianity due to the various astrophysical effects such as the Wouthysen-Field (WF) effect, X-ray heating and reionization. We study the various properties of 21cm bispectrum such as scale dependence, shape dependence and redshift evolution. And also we study the contribution from each component of 21cm bispectrum. We find that the contribution from each component has characteristic scale-dependent feature. In particular, we find that the bulk of the 21cm bispectrum at z =20 comes from the matter fluctuations, while in other epochs it is mainly determined by the spin and/or neutral fraction fluctuations and it is expected that we could obtain more detailed information on the IGM in the CD and EoR by using the 21cm bispectrum in the future experiments, combined with the power spectrum and skewness.

We perform a Fisher analysis to estimate the expected constraints on the Epoch of Reionization (EoR) model parameters (i.e., minimum virial temperature, the ionizing efficiency and the mean free path of ionizing photons) taking into account the thermal noise of the existing telescopes, MWA and LOFAR. We consider how the inclusion of the 21cm bispectrum improves the constraints compared to using the power spectrum alone. Assuming that we perfectly remove the foregrounds, we found that the bispectrum, which is calculated by the 21cmFAST code, can constrain the EoR model parameters more tightly than the power spectrum since the bispectrum is more sensitive to the EoR model parameters than the power spectrum. We also found that degeneracy between the EoR model parameters can be reduced by combining the bispectrum with the power spectrum.