A novel stacked hybrid autoencoder for imputing LISA data gaps

TL;DR

This paper presents a novel deep learning model combining a denoising autoencoder and a recurrent neural network to accurately impute data gaps in LISA gravitational wave signals, improving data continuity and analysis accuracy.

Contribution

The paper introduces a stacked hybrid autoencoder model specifically designed for LISA data gap imputation, integrating feature extraction and temporal dynamics capture in a novel way.

Findings

Achieves over 99.97% overlap when gaps are outside merger phase

Achieves over 99% overlap when gaps occur during merger phase

Data gaps during merger cause biased parameter estimates, indicating need for protected periods

Abstract

The Laser Interferometer Space Antenna (LISA) data stream will contain gaps with missing or unusable data due to antenna repointing, orbital corrections, instrument malfunctions, and unknown random processes. We introduce a new deep learning model to impute data gaps in the LISA data stream. The stacked hybrid autoencoder combines a denoising convolutional autoencoder (DCAE) with a bi-directional gated recurrent unit (BiGRU). The DCAE is used to extract relevant features in the corrupted data, while the BiGRU captures the temporal dynamics of the gravitational-wave signals. We show for a massive black hole binary signal, corrupted by data gaps of various numbers and duration, that we yield an overlap of greater than 99.97% when the gaps do not occur in the merging phase and greater than 99% when the gaps do occur in the merging phase. However, if data gaps occur during merger time, we show that we get biased astrophysical parameter estimates, highlighting the need for "protected periods," where antenna repointing does not occur during the predicted merger time.