Blazar 是光变最为突出的 AGN，其光变模式非常丰富，不同源模式不同，相同源其光变可以出现模式变化，在不同波段上的光变差异显著。从单个波段光变的时标看，时标可以从分钟到十几年，跨越多个量级。这些光变是否有一致的原因？是什么机制造成了光变的丰富多样性？Fermi和RXTE卫星的高能光变监测数据，具有一致的采样间隔和长期的连续性，可以用来讨论这些光变的共性和差异，再结合光学和射电波段的光变数据，进一步与AGN 内在的物理过程和几何结构相联系。本项目将评估各种线性和非线性时间序列分析方法，采用一致的、有效的统计分析方法，对这些数据进行整体分析，寻找决定光变模式的内在参数，再对多波段光变数据进行自相关和互相关分析，研究这些参数在不同波段上的变化，并结合有关的AGN模型进行讨论。本项目还将寻找光变模式中出现整体转换的源，研究其模式变化的内在原因。

Blazars are AGNs with ultra violent variability and diverse variability patterns. Different sources usually show distinct patterns. Meanwhile, the patterns can be time dependent and significantly different at multi wavelengths in an individual source. The variability time scale spans many orders of magnitude, varying from minutes to decades. The long-standing and challenging questions are: is there any common mechanism that acts for blazar variabilities, and what make the patterns so rich? Owning to the consistency in data sampling and the continuity in long-term monitoring, the light curves obtained with Fermi and RXTE satellites are very useful to analyze the time domain characteristics of blazars. Combined with optical and radio light curves, the multi-wavelength data will be helpful to probe the underlying physics and geometry of AGNs. In this project, through evaluating various kinds of linear and non-linear methods on time sequence analysis, we will select consistent and effective statistical methods to analyze the multi-wavelength light curves. We will constrain the intrinsic parameters that are responsible to diverse variability patterns, and then study the frequency evolution of the parameters by autocorrelating and crosscorrelating the multi-wavelength variability data. We will also search for the sources with the nature of pattern transition at multi wavelenghts. The results will be applied to develope the theoretical models of blazars and AGNs.