Application of 3D U-Net Neural Networks in Extracting the Epoch of Reionization Signal from SKA-Low Observations Based on Real Observations of NCP Field from LOFAR
| Autor: | Gao, Li-Yang, Koopmans, Léon V. E., Mertens, Florent G., Munshi, Satyapan, Li, Yichao, Brackenhoff, Stefanie A., Ceccotti, Emilio, Chege, J. Kariuki, Acharya, Anshuman, Ghara, Raghunath, Giri, Sambit K., Iliev, Ilian T., Mellema, Garrelt, Zhang, Xin |
|---|---|
| Rok vydání: | 2024 |
| Předmět: | |
| Druh dokumentu: | Working Paper |
| Popis: | Neutral hydrogen serves as a crucial probe for the Cosmic Dawn and the Epoch of Reionization (EoR). Actual observations of the 21-cm signal often encounter challenges such as thermal noise and various systematic effects. To overcome these challenges, we simulate SKA-Low-depth images and process them with a deep learning method. We utilized foreground residuals acquired by LOFAR during actual North Celestial Pole field observations, thermal and excess variances calculated via Gaussian process regression, and 21-cm signals generated with 21cmFAST for signal extraction tests. Our approach to overcome these foreground, thermal noise, and excess variance components employs a 3D U-Net neural network architecture for image analysis. When considering thermal noise corresponding to 1400 hours of integration, U-Net provides reliable 2D power spectrum predictions, and robustness tests ensure that we get realistic EoR signals. Adding foreground residuals, however, causes inconsistencies below the horizon delay-line. Lastly, evaluating both thermal and excess variance with observations up to 3700 and 14000 hours ensures reliable power spectrum estimations within the EoR window and across nearly all scales, respectively. Comment: 18 pages, 17 figures |
| Databáze: | arXiv |
| Externí odkaz: |
|