Göttingen 2025 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 4: Simulation and Workflows

AKPIK 4.5: Vortrag

Freitag, 4. April 2025, 10:00–10:15, Theo 0.134

Simulating Polarisation in Radio Interferometry Experiments Using pyvisgen — •Anno Knierim, Christian Arauner, and Kevin Schmitz — TU Dortmund University, Dortmund, Germany

Recent approaches in radio astronomy aim to improve image cleaning in radio interferometry measurements using machine learning techniques. Reconstructing sources using these novel techniques has the advantage of being agnostic to initial parameters used in traditional cleaning algorithms.

The radionets project is a deep-learning framework developed at TU Dortmund University. The goal is to reconstruct calibrated observations with convolutional neural networks to produce high-resolution images. Deep learning approaches such as radionets require large amounts of training and validation data. One approach to simulating the required datasets is provided by the simulation tool pyvisgen.

pyvisgen utilises the Radio Interferometer Measurement Equation (RIME) to represent the measurement process of a radio interferometer. It produces images suitable as input to train deep-learning-based cleaning approaches. This talk presents the recent implementation of polarisation effects on radio waves.

Keywords: Radio astronomy; Radio interferometry; Machine learning; Simulations