Complex-amplitude noise and accuracy analysis of quantized analog optical processors
Within this dissertation, we estabHsh the accuracy limitations of hybrid optoelectronic processing systems based upon an empirical understanding of elemental device noise characteristics. We approach this investigation in three phases. In the first phase we develop an analytical understanding of the information-bearing signal space for discrete numeric processors and establish how generalized spatiotemporal noise processes introduce uncertainty into this signal space, resulting in accuracy limitations. The second phase estabHshes an empirical understanding of the intensity-based noise processes for common sources, modulators, and detectors used to construct these processing systems. The third phase develops an understanding, both analytically and empirically, of the complex-amplimde signal and noise characteristics of a variety of Hquid crystal spatial tight modulators. The investigation concludes with an interpretation of the accuracy implications resulting from the observed electrical, intensity, and complex-amplitude noise processes, and proposes techniques which may achieve accuracy enhancements in such systems.