Abstract

We propose and demonstrate an effective two-stage modified signed-digit optical computing technique (in contrast to previous three-stage techniques) based on spatial data encoding, polarization multiplexing, and multiple imaging. Our proposed reduction in operation stages requires a reference operation in addition to the transformation and weight operations common to three-stage systems. In our system's first stage a transformation (or weight) operation and a reference operation are implemented in parallel by use of four distinct polarization-multiplexed kernel operations. In the second stage the final desired result (e.g., addition and subtraction) and its complement are obtained in parallel with a single kernel operation. The operation speed of our two-stage modified signed-digit computing method is 33% faster than previous three-stage modified signed-digit algorithms.

© 1995 Optical Society of America

Modified-signed-digit arithmetic for multi-input digital optical computing

Shaomin Zhou, Scott Campbell, Weishu Wu, Pochi Yeh, and Hua-Kuang Liu
Appl. Opt. 33(8) 1507-1516 (1994)

Conditional symbolic modified signed-digit arithmetic using optical content-addressable memory logic elements

Yao Li and George Eichmann
Appl. Opt. 26(12) 2328-2333 (1987)

Compact parallel optical modified-signed-digit arithmetic-logic array processor with electron-trapping device

Guoqiang Li, Feng Qian, Hao Ruan, and Liren Liu
Appl. Opt. 38(23) 5039-5045 (1999)

Modified signed-digit trinary arithmetic by using optical symbolic substitution

A. A. S. Awwal, M. N. Islam, and M. A. Karim
Appl. Opt. 31(11) 1687-1694 (1992)

Two-step digit-set-restricted modified signed-digit addition–subtraction algorithm and its optoelectronic implementation

Feng Qian, Guoqiang Li, Hao Ruan, Hongmei Jing, and Liren Liu
Appl. Opt. 38(26) 5621-5630 (1999)