A NEW METHOD OF SPECTRAL-ANALYSIS OF UNEVENLY SPACED ASTROPHYSICAL DATA .1. THE SINGLE-PERIOD CASE

In this paper we present a new method for the Fourier spectral analysis of periodic signals, which takes into full account the irregular data sampling. It can also be applied to evenly spaced data, even if with less computational efficiency with respect to other methods, but with results independent of the sampling frequency. In the present implementation the method requires the knowledge of the period of the signal to analyze, but it will be soon extended to become a general spectral analysis method. In this paper we describe the method, showing its efficiency in solving some computer-generated test functions with different signal-to-noise ratios. Then we report the results of the analysis of three classic Cepheids (eta Aql, X Cyg, and U Aql) and of two type a RR Lyrae variables (SW And and TU UMa) for which light curves in different colors and radial velocity curves were available from the literature. All the solutions obtained are then statistically analyzed.