Optical Computing Strategies.
Abstract
The subject of the current research effort is the development of a theory of optical computing, it is generally agreed that optical computing has an advantage over digital computing in situations where parallelism can be exploited. The canonical examples are matrix vector multiplication and matrix-matrix multiplication. If the matrices are both square and of size (n x n) then outer-product decomposition achieves a saving in computational time because the N2 inner products can be evaluated concurrently. It is outlined in Section 1. Our second completed contribution is the development of a tractable mathematical model of an optical system (assuming incoherent light operations and its use into an investigation of the inherent limits of computation of such a system in terms of a lower bound on the simultaneous resources of volume and computing time. Keywords: Electrooptical, Matrix vector multiplication, Canonical.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 14, 1985
- Accession Number
- ADA175018
Entities
People
- Richard Barakat