Abstract Details
| Name: Madhurima Choudhury Affiliation: IIT, Indore Conference ID: ASI2019_147 Title : Extracting the 21cm Global Signal using artificial neural networks Authors and Co-Authors : Madhurima Choudhury, Abhirup Datta Abstract Type : Oral Abstract Category : General Relativity and Cosmology Abstract : Observations of HI 21 cm transition line would be an important and promising probe into the cosmic Dark Ages and Epoch of Reionization. Detection of this redshifted 21 cm signal is one of the key science goals for several upcoming and future low-frequency radio telescopes like Hydrogen Epoch of Reionization Array (HERA), Square Kilometer Array (SKA) and Dark Ages Radio Explorer (DARE). One of the major challenges for the detection of this signal is the accuracy of the foreground source removal. Several novel techniques have been explored already to remove bright foregrounds from both interferometric as well as total power experiments. Here, we present preliminary results from our investigation on the application of Artificial Neural Networks to detect faint 21cm global signal amidst the sea of bright galactic foreground. Following the formalism of representing the global 21cm signal by the tanh model (Mirocha et al. 2015), this study finds that the global 21cm signal parameters can be accurately determined even in the presence of bright foregrounds represented by 3rd order log-polynomial (Harker 2015) or higher. This presentation also deals with results of foreground removal in presence of instrumental noise. We have also dealt with extracting the signal when the parametric form of the signal is not taken into consideration, and have got some interesting results. |