Order-N sparse minimum-variance open-loop reconstructor for extreme adaptive optics.

摘要: A scalable sparse minimum-variance open-loop wave-front reconstructor for extreme adaptive optics (ExAO) systems is presented. The based on Ellerbroek’s approximation of the inverse covariance matrix [J. Opt. Soc. Am. 19, 1803 (2002)]. baseline numerical approach an iterative conjugate gradient (CG) algorithm reconstructing a spatially sampled wave front at N grid points computational domain size equal to telescope’s primary mirror’s diameter D that uses multigrid (MG) accelerator speed up convergence efficiently and enhance its robustness. combined MGCG scheme order requires only two CG iterations converge asymptotic average Strehl ratio (SR) root-mean-square reconstruction error. SR squared error are within standard deviation with figures obtained from previously proposed fast-Fourier-transform incorporates exact 2D . cost comparison between present Cholesky factorization reordering preserve sparsity indicates latter method still competitive real-time ExAO ∼ 104 degrees freedom because update rate factor typically several orders magnitude lower than temporal sampling rate.