| Adaptive Optics (AO) is a method of correcting the distortion in astronomical images in real-time as the images are formed. AO systems are becoming more complicated with more elements to measure and correct the atmospheric distortion, faster sampling rates, and larger telescope diameters. It has become increasingly more important to simulate these systems before they are constructed to estimate their performance. An important part of the simulators is the simulation of the distorting atmosphere. Four algorithms for simulating the distorting effects of the atmosphere are investigated. These algorithms are spectrally synthesized fractional Brownian motion (SSFBM), discrete FBM (dFBM), successive random addition (SRA), and fractional Gaussian processes (FGP). These algorithms are used to simulate distortions following the von Karman model and the FGP algorithm is found to work the best at simulating data with the desired correlations and power spectra. There are four commonly used models of the spatial spectrum of atmospheric phase distortions: Kolmogorov, von Karman, exponential and Greenwood. This work introduces a new model, called the zero-low-frequency model, for the spatial spectrum of wavefront phase distortion. The von Karman, exponential, Greenwood and zero-low-frequency models include a parameter called the outer scale of turbulence used to model the largest turbulent eddy size in the atmosphere. The temporal spectrum of wavefront slope data from a wavefront measuring device has a roll-over frequency proportional to the outer scale where the slope changes from positive to negative on a log-log graph. A nonlinear modelling technique, called iterative robust orthogonal search (ITROS), is used to estimate the location of this roll-over frequency from experimental wavefront slope data. The estimate of the roll-over frequency is used to estimate the outer scale from two observation sites. Lastly, the theoretical effects of the zero-low-frequency model on the wavefront distortions are explored. |
