Abstract

Optical computers can be assembled by using fixed-interconnect structures to connect two-dimensional optical arrays. A four-phase model of the optical computer assembly is described, and the roles of design density (i.e., the number of available gates used in a design) and fault density and the assembly technique are empirically investigated. Low-cost array assignment techniques that make use of optical logic arrays with faulted gates and multiple-array orientations are introduced and shown to increase an effective yield by between 50 and 100%. These techniques provide near-optimal solutions to the general optimal array assignment problem, which is shown to belong to the complexity class NP-hard.

© 1992 Optical Society of America

Design and analysis of fixed planar holographic interconnects for optical neural networks

Paul E. Keller and Arthur F. Gmitro
Appl. Opt. 31(26) 5517-5526 (1992)

Implications of interconnection theory for optical digital computing

Haldun M. Ozaktas and Joseph W. Goodman
Appl. Opt. 31(26) 5559-5567 (1992)

Bit-serial architecture for optical computing

Vincent P. Heuring, Harry F. Jordan, and Jonathan P. Pratt
Appl. Opt. 31(17) 3213-3224 (1992)

Complexity analysis of optical-computing paradigms

Ahmed Louri and Arthur Post
Appl. Opt. 31(26) 5568-5583 (1992)

Free-space photonic switching architectures based on extended generalized shuffle networks

Thomas J. Cloonan, Gaylord W. Richards, Anthony L. Lentine, Frederick B. McCormick, and John R. Erickson
Appl. Opt. 31(35) 7471-7492 (1992)