Analysis and Synthesis of Three-dimensional Illumination Using Partial Coherence a Dissertation Submitted to the Department of Electrical Engineering and the Committee on Graduate Studies of Stanford University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

There exists many devices that generate three-dimensional illumination patterns. Analysis of the capabilities of three major device families using measures of partial coherence reveals specific illumination patterns that cannot be generated by each device family. Ray-based devices cannot achieve patterns with high resolution, coherent holographic devices cannot achieve certain intensity patterns, even in the twodimensional case, and volumetric devices cannot simulate occlusions and suffer from out-of-focus blur. Synthesis of more versatile illumination patterns is proposed by computing the mutual intensity representation of a desired partially coherent beam from application specifications and then generating the beam using time-multiplexing methods based on coherent modes. The mutual intensity can be computed directly from a simple scene description, or it can be computed through a novel algorithm using nonlinear conjugate gradients from a desired three-dimensional intensity volume. Equivalent coherent mode decomposition representations for the same mutual intensity will be considered in terms of optimality in efficiency. For cases when the computed mutual intensity is fairly incoherent, a new “quasi-Schell” mode decomposition is proposed to reduce the number of patterns needed at the SLM by introducing partially coherent sources. Use of arbitrary partially coherent beams for three-dimensional illumination enables the versatile generation of light patterns not possible with current devices and is a promising new field for exploration.