Speaker
Description
Extreme Mass-Ratio Inspirals (EMRIs) are among the targeted gravitational wave sources for LISA, promising valuable insights into fundamental physics and astrophysics. Preparing for the complex task of EMRI data analysis requires realistic, synthetic datasets. This talk presents the EMRIs for the Mojito light dataset, the upcoming Mock Data Challenge by the LISA DDPC designed for LISA data analysis studies. The catalog is constructed based on an astrophysical population model, which predicts the distribution of EMRI sources throughout the universe. Waveforms are generated using the latest waveform model in the FastEMRIWaveforms package, producing relativistic waveforms for eccentric, equatorial inspirals into spinning massive black holes. To validate the dataset and assess the fidelity of the waveform model, we perform a parameter reconstruction analysis. Using Bayesian inference with MCMC sampling, we quantify the accuracy with which source parameters can be recovered from the simulated data. The analysis is performed with and without noise, to test the performance of the new waveform model. Also, an estimate is made on the impact of overlapping signals on parameter reconstruction. Our results demonstrate the robustness of the waveform generator and provide estimates for the accuracy that can be expected for the EMRIs in this dataset. We find that the overlap has negligible impact for on the parameter reconstruction.
