Date of Award


Document Type


Degree Name

Master of Science


Department of Electrical and Computer Engineering

First Advisor

Eric P. Magee, PhD


Branch point tolerant phase reconstructors can vastly improve adaptive optic system performance in extended atmospheric turbulence. This thesis explores the performance bounds of two such reconstructors Goldstein's algorithm and hidden phase. A least squares reconstructor is implemented for comparison. System performance is presented for various scenarios, including correction time-delays, wave-front sensor noise, and extended beacons. These scenarios are of interest for laser communication and directed energy systems such as Airborne Laser. Performance bounds are obtained through wave-optics simulation. The extended beacon propagation geometry approximates the USAF AFRL-DE North Oscura Peak range. Results show that branch point tolerant reconstructors outperform least squares for equal correction time-delays. These reconstructors can be made somewhat tolerant to wave-front sensor error. For the case of an incoherent extended beacon, branch point information is lost and the branch point algorithms perform on par with least squares. A coherent extended beacon preserves branch point information, but also induces branch point errors due to coherent speckle. Still, the branch point reconstructors tend to maintain a 1-2 order of magnitude performance advantage over least squares in strong turbulence. While implementation challenges remain, this thesis demonstrates the potential of branch point tolerant phase reconstructors on laser communication and weapons systems.

AFIT Designator


DTIC Accession Number