Automatic search for periodic light curves with any kind of shape in unevenly spaced data.
Résumé
It is shown that the "periodogram analysis" commonly used to find sinusoid-like periodicities can be improved to search for variable star light curves of any kind of shape. The use of a particular set of orthogonal functions for the Least Square Model fitted at different test frequencies permits the derivation of a generalization of the periodogram and a probability of detection better adapted to the detection of any sort of periodic light curve. The results of Monte Carlo simulations are presented to illustrate the advantage of the new method.
Origine : Fichiers éditeurs autorisés sur une archive ouverte