|M.Sc Student||Paykin Irina|
|Subject||Simulated Annealing of Multiple-Aperture Telescopes|
|Department||Department of Physics||Supervisors||Dr. Joan Adler|
|Dr. Erez Ribak|
|Full Thesis text|
A crucial part of segmented or multiple-aperture systems is control of the optical path difference between the segments or sub-apertures. In order to achieve optimal performance we have to phase sub-apertures to within a fraction of the wavelength, and this requires high accuracy of positioning for each sub-aperture. We present the first simulations of a simulated annealing algorithm which are carried out in tandem with a hardware realization of a simulated annealing algorithm by E. N. Ribak and L. Yacobi in an active optical system with sparse segments. In order to align the optical system we applied the optimization algorithm to the image itself. The main advantage of this method over traditional correction methods is that wave-front sensing hardware and software are no longer required, making the optical and mechanical system much simpler.
The results of simulations and laboratory experiments demonstrate the ability of this optimization algorithm to correct both piston and tip/tilt errors.In addition we explored image restoration techniques, required for sparse aperture systems. Several classic deconvolution algorithms, such as Wiener-Helstrom, Lucy-Richardson and blind deconvolution, were implemented and investigated. Most deconvolution methods require prior knowledge of the practical point spread function (PSF), but blind deconvolution provides the possibility to reconstruct an image without any prior knowledge of the PSF. Finally, diffraction and aberration effects related to specific pupil configurations of sparse aperture telescopes were analyzed.