Rodrigo Tenorio (UNIMIB/INFN/UIB) — Searching for continuous gravitational-wave signals from unknown sources in Advanced LIGO O4 data

💡A RES Success Story about uncovering the unseen neutron star population with GPU-accelerated algorithms💡

📋"Searching for continuous gravitational-wave signals from unknown sources in Advanced LIGO O4 data" led by Rodrigo Tenorio (Università degli Studi di Milano-Bicocca/INFN/Universitat de les Illes Balears)

Continuous gravitational waves (CGWs) are faint, long-duration signals from rapidly spinning neutron stars, which detection requires searches over vast parameter-space regions that conventional algorithms cannot cover at scale. To address this challenge they have developed FastTracks, a novel data-analysis pipeline designed to run efficiently on modern GPU architectures.

This method avoids some approximations and achieves a 20% boost in sensitivity, allowing to survey a 70% more volume with the same computational cost. They combined it with a semi-analytical sensitivity estimation algorithm to predict the search reach, allowing to make the entire analysis cycle more efficient.

🖥️Thanks to RES supercomputer #MareNostrum5 ACC at Barcelona Supercomputing Center, the team executed a full-scale all-sky search in data from the fourth observing run (O4) of the LIGO Laboratory detectors.

These computations probed a large portion of the neutron star binary parameter space, including systems with significant spin-wandering due to accretion. These are promising but currently undetected sources of CGWs that can now be explored, paving the way to allow the a potential first-ever detection of a CGW signal.

FastTracks and supporting open-source tools (https://lnkd.in/d6c2xMrP) are available for the community and are expected to play a central role in upcoming searches with both ground-based and space-based detectors:

📸 The images show an example of CGWs source in a binary system, as well as the efficiency of FastTracks for different pipeline configurations.