A transformer-based deep learning network reconstructs VLBI images directly from closure invariants, achieving median NXCORR fidelity above 0.9 on untrained synthetic morphologies.
A VLBI receiving system for the South Pole Telescope
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
The Event Horizon Telescope (EHT) is a very-long-baseline interferometry (VLBI) experiment that aims to observe supermassive black holes with an angular resolution that is comparable to the event horizon scale. The South Pole occupies an important position in the array, greatly increasing its north-south extent and therefore its resolution. The South Pole Telescope (SPT) is a 10-meter diameter, millimeter-wavelength telescope equipped for bolometric observations of the cosmic microwave background. To enable VLBI observations with the SPT we have constructed a coherent signal chain suitable for the South Pole environment. The dual-frequency receiver incorporates state-of-the-art SIS mixers and is installed in the SPT receiver cabin. The VLBI signal chain also includes a recording system and reference frequency generator tied to a hydrogen maser. Here we describe the SPT VLBI system design in detail and present both the lab measurements and on-sky results.
fields
astro-ph.IM 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Deep Learning VLBI Image Reconstruction with Closure Invariants
A transformer-based deep learning network reconstructs VLBI images directly from closure invariants, achieving median NXCORR fidelity above 0.9 on untrained synthetic morphologies.