Particle Swarm Optimization based search for gravitational waves from compact binary coalescences: performance improvements

TL;DR

This paper demonstrates that Particle Swarm Optimization significantly enhances the efficiency of gravitational wave searches from compact binary coalescences across second-generation detectors, reducing computational costs while maintaining effectiveness.

Contribution

The study extends PSO-based search methods to second-generation detectors and introduces improvements like local-best PSO and optimized parameter tuning for better performance.

Findings

PSO reduces likelihood evaluations by a factor of 10 compared to grid-based methods.

The method is effective across the entire relevant binary mass range.

Performance improvements enable viable all-sky searches with second-generation detectors.

Abstract

While a fully-coherent all-sky search is known to be optimal for detecting signals from compact binary coalescences (CBCs), its high computational cost has limited current searches to less sensitive coincidence-based schemes. For a network of first generation GW detectors, it has been demonstrated that Particle Swarm Optimization (PSO) can reduce the computational cost of this search, in terms of the number of likelihood evaluations, by a factor of $\approx 10$ compared to a grid-based optimizer. Here, we extend the PSO-based search to a network of second generation detectors and present further substantial improvements in its performance by adopting the local-best variant of PSO and an effective strategy for tuning its configuration parameters. It is shown that a PSO-based search is viable over the entire binary mass range relevant to second generation detectors at realistic signal strengths.