The Application of Continuous Wavelet Transform Based Foreground Subtraction Method in 21 cm Sky Surveys

TL;DR

This paper introduces a wavelet transform-based non-parametric method for foreground subtraction in 21 cm sky surveys, effectively distinguishing signals from complex foregrounds and instrument errors with less computational cost.

Contribution

The paper presents a novel wavelet transform approach that improves foreground subtraction accuracy and efficiency over traditional spectral fitting methods in 21 cm surveys.

Findings

Effective in separating 21 cm signals from complex foregrounds.

More tolerant to instrument response errors than traditional methods.

Consumes less computational time than similar non-parametric methods.

Abstract

We propose a continuous wavelet transform based non-parametric foreground subtraction method for the detection of redshifted 21 cm signal from the epoch of reionization. This method works based on the assumption that the foreground spectra are smooth in frequency domain, while the 21 cm signal spectrum is full of saw-tooth-like structures, thus their characteristic scales are significantly different. We can distinguish them in the wavelet coefficient space easily and perform the foreground subtraction. Compared with the traditional spectral fitting based method, our method is more tolerant to complex foregrounds. Furthermore, we also find that when the instrument has uncorrected response error, our method can also work significantly better than the spectral fitting based method. Our method can obtain similar results with the Wp smoothing method, which is also a non-parametric method, but our method consumes much less computing time.