LIGO Document T1600288-v2

Improving the stochastic template bank algorithm used for detection of compact binary systems by Advanced LIGO

Document #:
Document type:
T - Technical notes
Other Versions:
Within the next few years, LIGO anticipates between tens and hundreds of gravitational wave (GW) detections. This increase in signal events will inevitably require a more efficient form of analysis. Extending the current method of providing template simulations for analysis with a more intelligent coverage of the parameter space might avoid excessive delays and computational costs in analysis. The Metropolis-Hastings algorithm generates intelligent proposal templates to increase the likelyhood of placement in the bank of templates. Exploiting the benefits of parallel computing on powerful multi-core machines offers the potential for dramatic improvements in runtime at no cost of coverage. Thus, current functions contained in the inspiral libraries will be re-factored in the interest of having a parallelized stochastic template bank generation before the start of Advanced LIGO's first observation run.

DCC Version 3.4.3, contact Document Database Administrators