Darwin Meets Einstein: LISA Data Analysis Using Genetic Algorithms

TL;DR

This paper demonstrates the application of Genetic Algorithms to analyze LISA gravitational wave data, effectively extracting parameters of overlapping sources in a complex, high-dimensional space.

Contribution

It introduces the first use of GAs for LISA data analysis, optimizing global search for multiple overlapping gravitational wave sources.

Findings

GAs successfully extract source parameters in simulated LISA data.

Optimizations improve the efficiency and accuracy of the GA approach.

GAs handle large, overlapping source signals effectively.

Abstract

This work presents the first application of the method of Genetic Algorithms (GAs) to data analysis for the Laser Interferometer Space Antenna (LISA). In the low frequency regime of the LISA band there are expected to be tens of thousands galactic binary systems that will be emitting gravitational waves detectable by LISA. The challenge of parameter extraction of such a large number of sources in the LISA data stream requires a search method that can efficiently explore the large parameter spaces involved. As signals of many of these sources will overlap, a global search method is desired. GAs represent such a global search method for parameter extraction of multiple overlapping sources in the LISA data stream. We find that GAs are able to correctly extract source parameters for overlapping sources. Several optimizations of a basic GA are presented with results derived from applications of the GA searches to simulated LISA data.