Foreground removal from WMAP 5yr temperature maps using an MLP neural network

TL;DR

This paper demonstrates that a multilayer perceptron neural network can effectively remove foreground contamination from WMAP 5-year temperature maps, achieving low systematic errors and high sky coverage, thus improving CMB signal extraction.

Contribution

The study applies a simple neural network to WMAP data without auxiliary inputs, showing improved foreground removal and error reduction over previous methods.

Findings

Over 75% sky coverage with low errors

Systematic errors due to foregrounds are minimal

Neural network method is ready for Planck data analysis

Abstract

One of the main obstacles for extracting the cosmic microwave background (CMB) signal from observations in the mm/sub-mm range is the foreground contamination by emission from Galactic component: mainly synchrotron, free-free, and thermal dust emission. The statistical nature of the intrinsic CMB signal makes it essential to minimize the systematic errors in the CMB temperature determinations. The feasibility of using simple neural networks to extract the CMB signal from detailed simulated data has already been demonstrated. Here, simple neural networks are applied to the WMAP 5yr temperature data without using any auxiliary data. A simple \emph{multilayer perceptron} neural network with two hidden layers provides temperature estimates over more than 75 per cent of the sky with random errors significantly below those previously extracted from these data. Also, the systematic errors, i.e.\ errors correlated with the Galactic foregrounds, are very small. With these results the neural network method is well prepared for dealing with the high - quality CMB data from the ESA Planck Surveyor satellite.