Acceleration of Computation Speed for Wavefront Phase Recovery Using Programmable Logic

Abstract

Atmospheric turbulence introduces optical aberration into wavefronts arriving at groundbased telescopes. Current adaptive optics (AO) systems use vector-matrix-multiply (VMM) reconstructors to convert gradient measurements to wavefront phase estimates. Until recently, the problem of an efficient phase recoverer design has been implemented over PC or GPU platforms. As the number of actuators n increases, the time to compute the reconstruction by means of the VMM method scales as O(n2). The number of actuators involved in AO systems is expected to increase dramatically in the future. For instance, the increase in the field of astronomy is due to increasing telescope diameters and new higherresolution applications on existing systems. The size increase ranges from hundreds up to tens of thousands of actuators and requires faster methods to complete the AO correction within the specified atmospheric characteristic time. The next generation of extremely large telescopes (with diameters measuring from 50 up to 100 meters) will demand important technological advances to maintain telescope segment alignment (phasing of segmented mirrors) and posterior atmospheric aberrations corrections. Furthermore, an increase in telescope size requires significant computational power. Adaptive optics includes several steps: detection, wavefront phase recovery, information transmission to the actuators and their mechanical movements. A quicker wavefront phase reconstruction appears to be an extremely relevant step in its improvement. For this reason other hardware technologies must be taken into account during the development of a specific processor.