Improving SEOBNRv3 for LIGO Parameter Estimation
ORAL
Abstract
The Spinning Effective One Body--Numerical Relativity (SEOBNR) series of gravitational wave approximants are among the best available for Advanced LIGO/Virgo parameter estimation (PE). Over the past two years we have sped up, via direct code optimization, the eight-dimensional SEOBNR version 3 (SEOBNRv3) approximant by a factor of ~340x. This is particularly significant since there is no known way to develop an efficient eight-dimensional Reduced-Order Model that is faster than the SEOB code itself. We report on our latest work, both in speeding up our optimized SEOBNRv3 approximant (SEOBNRv3_opt) by another order of magnitude and in making SEOBNRv3_opt more robust than SEOBNRv3. This work brings direct SEOBNRv3-based PE within reach for many important LIGO sources.
*This work was supported primarily by NSF LIGO Research Support Grant PHY-1607405, with additional funding from NSF EPSCoR Grant 1458952, and NASA Grants 13-ATP13-0077 and 80NSSC18K0538.
–
Presenters
-
Tyler Knowles
- West Virginia University