Title :
Multiple periodic oscillations analysis on astronomy signals using an Empirical Mode Decomposition-Ornstein Uhlenbeck method
Author :
Dian-Qiao Dong;Jun-Yi Wang;Tao An;Hong-Bing Qiu;Xiang-Long Lu
Author_Institution :
Key Laboratory of Cognitive Radio & Information Processing, the Ministry of Education, Guilin University of Electronic Technology, Guilin 541004, China
Abstract :
In this paper, we propose an Empirical Mode Decomposition-Ornstein Uhlenbeck (EMD-OU) method, by which we can investigate X-ray Binary (XB) variability. The Ornstein Uhlenbeck (OU) process can homogenize the source based on its own features. The method was used to verify eight high and low mass X-ray Binary data derived from Swift/BAT Website from 2005 to 2013. The results confirm that: The objects we studied show multiple periodicity features and the periodic variability components have simply proportion relationship with each other. The period of SS433 investigated by the EMD-OU method is 166 days, which is consistent to 164 days by the previous work. The EMD-OU method was verified to be effective. In addition, the periodic components obtained by our method are the same with those obtained by using Lomb-Scargle (LS) periodogram, which is developed for unevenly sampled source in astronomy.
Keywords :
"Time series analysis","Transforms","Time-frequency analysis","Oscillators","Astronomy","Stochastic processes"
Conference_Titel :
Image and Signal Processing (CISP), 2015 8th International Congress on
DOI :
10.1109/CISP.2015.7408059