Abstract

Optical array logic (OAL) is proposed as a basis of parallel optical digital computing.¹ This technique is applied to an optical digital computing system called OPALS (optical parallel array logic system).² Processing of the OPALS is based on SIMD (single instruction multiple data) and iterative structures, which is designed to utilize the characteristics of OAL effectively. However, there are problems related to the structure of the OPALS. Since the OPALS has a closed architecture, it is difficult to extend the system. Although many data can be processed simultaneously, minute operations for a part of data cause processing inefficiency. In addition, necessity of three dynamic components in the OPALS results in complication of hardware and degradation of processing throughput. Thus a new concept for system construction is an important subject in OAL scheme.

© 1991 Optical Society of America