GERBLS—A Rapid Box Least Squares Algorithm for Exoplanet Detection That Scales Well for Large Datasets
Year: 2026
Type: article
Abstract: Box least squares (BLS) is a robust but computationally expensive method to search for transits in photometric data. Reducing its computational cost is especially relevant when scanning large datasets or conducting transit injection and recovery simulations. Greatly Expedited Revamp of BLS (GERBLS) is a software algorithm designed to substantially speed up a BLS search. It offers O ( n log n ) runtime scaling as opposed to O ( n 2 ) for native brute-force BLS. GERBLS is based on a divide-and-conquer FFA that minimizes mathematically redundant operations during phase-folding. It can automatically downsample the input data at longer tested periods and yield a BLS period resolution that is close to optimal for transiting planets on circular orbits. GERBLS typically outperforms an efficient CPU-based brute-force BLS implementation by at least 10–100 times in total runtime, with a relative speedup of 600× demonstrated for a dataset consisting of 10 5 observations.
Source: The Astronomical Journal
Language: English
Subfield: Astronomy and Astrophysics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $4,499
OpenAlex ID: W7220674479