Time series analysis of gravitational wave signals using neural networks

10th Mexican School on Gravitation and Mathematical Physics: "Reaching a Century: Classical and Modified General Relativity's Attempts to explain de evolution of the Universe"'

物理学;数学

Carrillo, M.^1 ; González, J.A.^1 ; Gracia-Linares, M.^1 ; Guzmán, F.S.^1

Instituto de Física y Matemáticas, Universidad Michoacana de San Nicolás de Hidalgo, Edificio C-3, Cd. Universitaria, Morelia, Michoacán

58040, Mexico^1

关键词: Black holes;    Gravitational-wave signals;    Mass ratio;    Small samples;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/654/1/012001/pdf
DOI  :  10.1088/1742-6596/654/1/012001

来源: IOP

【 摘 要 】

We present an algorithm based on neural networks that predicts the mass ratio in a binary black hole collision out of given Gravitational Wave (GW) signals. In this brief analysis, the network is trained with a small sample of GW signals generated with numerical simulations. The effectiveness of the algorithm is evaluated with GW generated again with simulations with given mass ratios unknown to the network and found to be in the worst scenarios less than 10%.