Erlangen 2026 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 19: Gravitational Waves I

T 19.2: Vortrag

Montag, 16. März 2026, 16:30–16:45, KS 00.005

Parameter Estimation for long duration Gravitational Wave signals at the Einstein Telescope using Deep Learning — •Tobias Reike¹, Johannes Erdmann¹, and Achim Stahl² — ¹III. Physikalisches Institut A, RWTH Aachen University — ²III. Physikalisches Institut B, RWTH Aachen University

The proposed Einstein Telescope will be a third-generation gravitational-wave detector, succeeding the current detectors LIGO, Virgo, and KAGRA. It aims to extend the sensitive frequency band toward both lower and higher frequencies and to improve the sensitivity of the current detectors by an order of magnitude. As a result, detected signals can be observed for much longer durations, ranging from minutes to hours, and the detection rate is expected to increase dramatically, reaching hundreds per day.

The analysis methods currently used to estimate source parameters from detected signals are extremely demanding in terms of computational resources, making them unsuitable for the substantially larger data volume anticipated for the Einstein Telescope. Consequently, new and more efficient methods are under development. We present a deep-learning-based approach to parameter estimation that relies on conditional normalizing flows, along with our ongoing work on the analysis of long-duration signals, which pose a particular challenge.

Keywords: Deep Learning; Gravitational Waves; Einstein Telescope; Normalizing Flows; Parameter Estimation