Citation
Abstract
Present adaptive optics systems use a wavefront sensor to detect phase errors in the incoming wavefront. Knowledge of these phase errors then is used to correct the incoming wavefront, reducing image distortion. However, these systems require that a portion of the incoming light be diverted to the wavefront sensor and away from the main receiver’s focal plane, which results in a loss of signal power in the receiver in optical communication applications. Accordingly, it is desirable to develop a system that relies entirely on the focal-plane intensity distribution as opposed to a separate wavefront sensor in order to detect and correct wavefront errors. Two common stochastic optimization techniques—genetic algorithms and gradient descent algorithms—are evaluated in this article.2 Although these algorithms are promising, further work is necessary to enable them to be used in practical adaptive optics systems due to their slow convergence speeds relative to the rate of change in the atmosphere.
Keywords
Details
- Volume
- 42-161
- Published
- May 15, 2005
- Pages
- 1–21
- File Size
- 331.0 KB