Extending Cosmic Ray Background in Space Experiments using Generative Adversarial Networks

Cosmic rays (CR) reaching telescope detectors in outer space are known<br> to induce glitches and background noise. The presence of CR noise significantly<br> influenced Cosmic Microwave Background (CMB) experiments, like Planck and<br> LiteBIRD, which have a long exposition and hard shelling or filtering. In order to<br> address this challenge, it is imperative to accurately simulate the CR background<br> throughout the duration of LiteBIRD’s three-year mission. However, state-of-the-art<br> Monte Carlo (MC) simulations are extremely computational expensive, typically<br> requiring 30 times the simulated period. We present the Cosmic Rays Artificial<br> Background (CRAB) code, extending MC simulations with Generative Adversarial<br> Networks (GAN). By leveraging GANs, we can efficiently generate a sufficient<br> number of genuine, statistically independent images, unlike traditional noise analysis<br> techniques combined with template expansion methods.