Multiobjective optimization for scattering mitigation and scattering screen reconstruction in very long baseline interferometry observations of the Galactic Center

Alejandro Mus;Ciriaco Goddi
2025-01-01

Abstract

Context. Imaging reconstruction of interferometric data is a hard and ill-posed inverse problem. Its difficulty is increased for observations of the Galactic Center, which is obscured by a scattering screen. This is because the scattering breaks the one-To-one correspondence between images and visibilities. Aims. Solving the scattering problem is one of the greatest challenges in radio imaging of the Galactic Center. We present a novel strategy for mitigating its effect and for constraining the screen itself using multiobjective optimization. Methods. We exploited the potential of evolutionary algorithms to describe the optimization landscape with the aim to recover the intrinsic source structure and the scattering screen that affects the data. Results. We successfully recovered the screen and the source in a wide range of simulated cases, including the speed of a moving screen at 230 GHz. Particularly, we recovered a ring structure in scattered data at 86 GHz. Conclusions. Our analysis demonstrates the huge potential that recent advancements in imaging and optimization algorithms offer in recovering image structures, even in weakly constrained, degenerated, and possibly multimodal settings. The successful reconstruction of the scattering screen opens the window to event-horizon scale works on the Galactic Center at 86 GHz up to 116 GHz and to the study of the scattering screen itself.
2025
Inglese
698
A299
Esperti anonimi
scientifica
Methods: numerical
Techniques: high angular resolution
Techniques: image processing
Techniques: interferometric
Mus, Alejandro; Toscano, Teresa; M??ller, Hendrik; Zhao, Guang-Yao; Lobanov, Andrei; Goddi, Ciriaco
1.1 Articolo in rivista
info:eu-repo/semantics/article
1 Contributo su Rivista::1.1 Articolo in rivista
262
6
open
Files in This Item:
File Size Format  
2025_Mus_aa53038-24.pdf

open access

Type: versione editoriale
Size 4.24 MB
Format Adobe PDF
4.24 MB Adobe PDF View/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Questionnaire and social

Share on:
Impostazioni cookie