Integrated optical guided-wave architectures for signal processing and computing

Integrated optical waveguides are used for distributing lightwaves throughout small, wideband dielectric circuits. Recently, optical waveguide architectures have been developed to accomplish a variety of complex signal processing and computing tasks. Despite this progress, there are still important areas in which the use of guided-waves either has not been well explored (ie. interferometry) or their use leads to architectures which suffer extreme limitations over their corresponding electronic implementations (ie. electro-optical analog-to-digital converters ADC). This dissertation explores some of these issues as well as develop new guided-wave architectures to address these types of problems.;Parallel architectures for the computation of a signal's higher-order domains are developed. These architectures provide the capability to detect in real-time, various orders of nonlinearities which might exist in the input signal to be analyzed (eg. quadratic phase coupling). These real-time architectures are easily adapted to accommodate a variety of spectral bandwidths and signal processing needs. The configurations also allow a degree of multifunctionality with the ability to interleave these different functions in real-time (eg. angle-of-arrival calculations: bispectral analysis: angle-of arrival calculations, etc.).;Also developed is a technical process (MRNS encoding) for obtaining a new class of electro-optical ADCs which have the capability of digitizing an input signal with very high resolution (;The substrate radiation modes which exist on these types of integrated optical circuits can adversely affect the performance of the guided-wave architectures. The propagating beam method of analysis is used to highlight some of these limitations and provide an estimation of the performance and required layout for the architectures developed in this dissertation.