Detecting LISA sources using time-frequency techniques

TL;DR

This paper explores the use of time-frequency spectrogram techniques to detect gravitational wave sources in LISA data, especially for challenging signals like EMRIs, as a computationally efficient initial detection method.

Contribution

It demonstrates that time-frequency methods can potentially identify the nearest EMRI events in LISA data, providing a promising first-stage detection approach.

Findings

Time-frequency techniques may detect the nearest EMRI events.

Initial results are promising using simplified data models.

The method offers a computationally cheap detection strategy.

Abstract

The planned Laser Interferometer Space Antenna (LISA) will detect gravitational wave signals from a wide range of sources. However, disentangling individual signals from the source-dominated data stream is a challenging problem and the focus of much current research. The problems are particularly acute for detection of extreme mass ratio inspirals (EMRIs), for which the instantaneous signal amplitude is an order of magnitude below the level of the instrumental noise, and the parameter space of possible signals is too large to permit fully-coherent matched filtering. One possible approach is to attempt to identify sources in a time-frequency spectrogram of the LISA data. This is a computationally cheap method that may be useful as a first stage in a hierarchical analysis. Initial results, evaluated using a significantly simplified model of the LISA data stream, suggest that time-frequency techniques might be able to detect the nearest few tens of EMRI events. In this proceedings article, we briefly outline the methods that have so far been applied to the problem, initial results and possible future directions for the research.