Test of Waveform Consistency with chirplets

Gravitational-wave (GW) signals can be reconstructed using two complementary approaches. The first employs waveform templates derived from General Relativity (GR) to estimate source parameters. The second is implemented via an algorithm called BayesWave, which reconstructs signals using basis functions without assuming a specific physical model, enabling robust tests of waveform consistency. Previous comparisons of the templated and template-free reconstructions have used a sine-Gaussian wavelet basis in BayesWave. However, allowing linear time evolution of the frequency in wavelets can lead to better reconstruction of low signal-to-noise ratio (SNR) events. We refer to these frequency-evolving wavelets as chirplets. In this work, we revisit the third Gravitational-Wave Transient Catalog using chirplet-based BayesWave reconstructions to perform updated waveform consistency tests. We first analyze all events using chirplets, and then restrict chirplet usage to low-SNR events. We compare both strategies to the standard wavelet-based analysis and quantify the improvement while using chirplets. Our results argue for the inclusion of chirplets in future consistency tests to broaden the range of signals that can be robustly analyzed.