21-cm foreground removal using AI and frequency-difference technique

Abstract

The deep learning technique has been employed in removing foreground contaminants from 21 cm intensity mapping, but its effectiveness is limited by the large dynamic range of the foreground amplitude. In this study, we develop a novel foreground removal technique grounded in U-Net networks. The essence of this technique lies in introducing an innovative data preprocessing step specifically, utilizing the temperature difference between neighboring frequency bands as input, which can substantially reduce the dynamic range of foreground amplitudes by approximately two orders of magnitude. This reduction proves to be highly advantageous for the U-Net foreground removal. We observe that the HI signal can be reliably recovered, as indicated by the cross-correlation power spectra showing unity agreement at the scale of $k<0.3h^{-1}$Mpc in the absence of instrumental effects. Moreover, accounting for the systematic beam effects, our reconstruction displays consistent auto-correlation and cross-correlation power spectrum ratios at the $1\sigma$ level across scales $k\lesssim 0.1h^{-1}$Mpc, with only a 10% reduction observed in the cross-correlation power spectrum at $k\simeq 0.2h^{-1}$Mpc. The effects of redshift-space distortion are also reconstructed successfully, as evidenced by the quadrupole power spectra matching. In comparison, our method outperforms the traditional Principal Component Analysis method, which derived cross-correlation ratios are underestimated by around 60%. We simulated various white noise levels in the map and found that the mean cross-correlation ratio ${\bar{R}}_{\mathrm{cross}}\gtrsim 0.8$ when the level of the thermal noise is smaller than or equal to that of the HI signal. We conclude that the proposed frequency-difference technique can significantly enhance network performance by reducing the amplitude range of foregrounds and aiding in the prevention of HI loss.

arXiv comment20 pages, 19 figures

References in corpus (61)

• HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
• A model of diffuse Galactic Radio Emission from 10 MHz to 100 GHz
• A unified pseudo-$C_{\ell }$ framework
• Baryon Acoustic Oscillation Intensity Mapping as a Test of Dark Energy
• Measuring the Small-Scale Power Spectrum of Cosmic Density Fluctuations Through 21 cm Tomography Prior to the Epoch of Structure Formation
• Measurement of 21 cm brightness fluctuations at z ~ 0.8 in cross-correlation
• Late-time cosmology with 21cm intensity mapping experiments
• Four Fundamental Foreground Power Spectrum Shapes for 21 cm Cosmology Observations
• Multifrequency analysis of 21 cm fluctuations from the Era of Reionization
• The Epoch of Reionization Window: I. Mathematical Formalism
• Foreground simulations for the LOFAR - Epoch of Reionization Experiment
• Determination of z~0.8 neutral hydrogen fluctuations using the 21 cm intensity mapping auto-correlation
• The pre-launch Planck Sky Model: a model of sky emission at submillimetre to centimetre wavelengths
• HI intensity mapping : a single dish approach
• Data Analysis for Precision 21 cm Cosmology
• Radio Foregrounds for the 21cm Tomography of the Neutral Intergalactic Medium at High Redshifts
• All-Sky Interferometry with Spherical Harmonic Transit Telescopes
• A Method for 21cm Power Spectrum Estimation in the Presence of Foregrounds
• Foreground Removal using FastICA: A Showcase of LOFAR-EoR
• Lack of clustering in low-redshift 21-cm intensity maps cross-correlated with 2dF galaxy densities
• HIRAX: A Probe of Dark Energy and Radio Transients
• Detecting the Rise and Fall of 21 cm Fluctuations with the Murchison Widefield Array
• Twenty-one centimeter tomography with foregrounds
• Neutral hydrogen surveys for high redshift galaxy clusters and proto-clusters
• The Scale of the Problem : Recovering Images of Reionization with GMCA
• Foregrounds for 21cm Observations of Neutral Gas at High Redshift
• Simulating polarized Galactic synchrotron emission at all frequencies, the Hammurabi code
• A ground-based 21cm Baryon acoustic oscillation survey
• Using HI to probe large scale structures at z ~ 3
• HI intensity mapping with MeerKAT: power spectrum detection in cross-correlation with WiggleZ galaxies
• HI constraints from the cross-correlation of eBOSS galaxies and Green Bank Telescope intensity maps
• Detection of Cosmological 21 cm Emission with the Canadian Hydrogen Intensity Mapping Experiment
• Forecasts on the Dark Energy and Primordial Non-Gaussianity Observations with the Tianlai Cylinder Array
• De-contamination of cosmological 21-cm maps
• Interpreting the unresolved intensity of cosmologically redshifted line radiation
• HI intensity mapping with MeerKAT: Calibration pipeline for multi-dish autocorrelation observations
• Fast simulations for intensity mapping experiments
• 21cm foregrounds and polarization leakage: a user's guide on cleaning and mitigation strategies
• Extracting HI cosmological signal with Generalized Needlet Internal Linear Combination
• deep21: a Deep Learning Method for 21cm Foreground Removal
• The 21cm Power Spectrum After Reionization
• The BINGO Project I: Baryon Acoustic Oscillations from Integrated Neutral Gas Observations
• Gaussian Process Foreground Subtraction and Power Spectrum Estimation for 21 cm Cosmology
• Hi intensity mapping with MeerKAT: Primary beam effects on foreground cleaning
• The Tianlai Dish Pathfinder Array: design, operation and performance of a prototype transit radio interferometer
• Tapering the sky response for angular power spectrum estimation from low-frequency radio-interferometric data
• The foreground transfer function for HI intensity mapping signal reconstruction: MeerKLASS and precision cosmology applications
• The cross-correlation between 21cm intensity mapping maps and the Lyman-alpha forest in the post-reionization era
• Bayesian semi-blind component separation for foreground removal in interferometric 21-cm observations
• Eliminating Primary Beam Effect in Foreground Subtraction of Neutral Hydrogen Intensity Mapping Survey with Deep Learning
• Measuring the X-ray Background in the Reionization Era with First Generation 21 cm Experiments
• Joint estimation of the Epoch of Reionization power spectrum and foregrounds
• A Bayesian analysis of redshifted 21-cm HI signal and foregrounds: Simulations for LOFAR
• The Tianlai dish array low-z surveys forecasts
• HIR4: cosmology from a simulated neutral hydrogen full sky using Horizon Run 4
• A Semi-blind PCA-based Foreground Subtraction Method for 21 cm Intensity Mapping
• Simulated foreground predictions for HI at z = 3.35 with the Ooty Wide Field Array: I. Instrument and the foregrounds
• Galactic Faraday rotation effect on polarization of 21 cm lines from the epoch of reionization
• All Sky Modelling Requirements for Bayesian 21 cm Power Spectrum Estimation with BayesEoR
• HIR4: Cosmological signatures imprinted on the cross correlation between 21-cm map and galaxy clustering
• The redshifted HI 21 cm signal from the post-reionization epoch: Cross-correlations with other cosmological probes

Cited by in corpus (4)

• DarkAI: Reconstructing the density, velocity and tidal field of dark matter from a DESI-like bright galaxy sample
• Deep learning approach for identification of HII regions during reionization in 21-cm observations -- III. image recovery
• Foreground removal in HI 21 cm intensity mapping under frequency-dependent beam distortions
• Deep learning with hybrid frequency differencing and principal component analysis for 21-cm foreground and beam mitigation