Application of Photorefractive Crystals to Solve Matrix Algebra Problems Optically
Abstract
The goals for the project are to solve matrix algebraic problems at high speed by applying photorefractive crystals in optical processing systems, and to study the photorefractives for such application. Although our interest was primarily on solving matrix algebraic problems, in order to improve system performance we found it important to optimize crystal operation and to improve the photorefractive memory performance. For example, a fast response and large coupling gain bandwidth photorefractive amplifier is required in the matrix inversion system, and high photorefractive efficiency and large storage capacity are required for the correlation matrix-tensor multiplier system. The report will be divided into four sections. Section 2.1 describes an all-optical implementation of an iterative algorithm for matrix inversion. Section 2.2 reports the implementation of a correlation matrix tensor multiplier algorithm in photorefractive crystals. Section 2.3 summaries the results that have been obtained in improving photorefractive performance, which includes 450-cut BaTiO3, applying electrical field and moving grating on SBN and GaP. Section 2.4 reviews our work on photorefractive memory, which includes circulating memory, incremental recording, orthogonal phase encoding and selective erasure. A list of publications resulted from the research is included at the end of this report.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 27, 1992
- Accession Number
- ADA259067
Entities
People
- Sing H. Lee
Organizations
- University of California, San Diego