Date of Award


Document Type


Degree Name

Master of Science in Electrical Engineering


Department of Electrical and Computer Engineering

First Advisor

Jason D. Schmidt, PhD


The Air Force Research Laboratory’s Sodium Guidestar Adaptive Optics for Space Situational Awareness program (NGAS) has sponsored research on spatially non-uniform gain for the servo-loop controller of an adaptive optics (AO) system. The edge subapertures of a Shack-Hartmann wavefront sensor have lower signal-to-noise ratios and are more susceptible to measurement errors than fully illuminated center subapertures. These measurement errors produce errant commands over the corresponding edge actuators and can induce instabilities over these regions in strong turbulence conditions. The objective of this research was to develop and experimentally verify the use of spatially varying gain maps on the servo-loop controller of a deformable mirror for improvements in the performance of an adaptive optics system and the mitigation of instabilities that can occur over the edge actuators of a deformable mirror. Spatially varying gain maps were developed to compensate for edge effects by having lower filter gains b designated for the edge subapertures. This lowered the bandwidth over the edge actuators which are the most susceptible to the edge effects. A gain map with a ring of lower filter gains b over just the outer actuators was experimentally shown to increase the overall Strehl ratio of the AO system in all of the tested turbulence conditions. Specifically, the non-uniform gain maps provide the largest increase of Strehl ratio over the AO system in low signal-to-noise ratio and high turbulent scenarios. Experiments were also conducted in regimes where instabilities formed over the edge actuators of the deformable mirror. Gaussian gain maps are shown to significantly reduce the overall residual phase variance over the edge actuators thereby reducing the formation of the instabilities.

AFIT Designator


DTIC Accession Number


Included in

Optics Commons