Using Lomb-Scargle Algorithm And Weighted Wavelet Z Transform Algorithm Analyze Periodicity Of X-ray Binaries

Variability of astronomical objects provides an efective tool to study theradiation mechanism and internal structure variation. In particular, periodicityhidden in the light curves of astronomical objects sheds light on the dynamic pro-cess tightly related to the central power source. The realistic astronomical lightcurves, owing to various reasons, are often irregularly sampled. The unevennessof the time series data often induces fake peaks in the power spectra, making itdifcult to identify genuine periodic signals. In some extreme cases, the noisyspectra lead to failture of periodicity identifcation. The present paper investi-gates the periodicity analysis methods of unevenly sampled time series, by takingthree famous X-ray binary as an example,namely SS433,HerX1and LMCX4. The Swift/BAT light curve of them was analyzed using the Lomb-Scargleperiodogram, and weighted wavelet z-transform methods. The wavelet transformis characterized of multi-resolution, varying scales and localized properties, andit improves the periodicity search and identifcation. The Monte Carlo tests showthat the statistical signifcance of their own period component is99.96%(>3.5σ),suggesting it is a genuine periodic signal. We also investigated how the noise af-fects the periodicity analysis. The weaker but longer periodic components seenin the wavelet spectra are probably generated by random process. The SS433periodicity detected from the X-ray light curve is consistent with that inferredfrom the radio light curve data. The periodic oscillations in the light curves ofSS433could be associated with the jet precession,the periodicity of HerX1andLMC X4may have connection with radiation-induced warping.