Hybrid Fuzzy-Color Computing Based on Optical Logical Architecture

Abstract

The article is devoted to developing an innovative approach to creating a high-speed optical computing architecture for processing a large amount of fuzzy information and supporting decision-making in artificial intelligence systems. The approach includes forming large fuzzy databases as a set of information quanta, and logical operations for inference solutions based on the principles of optical colorimetry, and fits well with the modern development of optical information transmission systems. The representation of an information color quantum as a digital fuzzy set allows the use of a well-developed fuzzy logic apparatus for a certain class of decision-making problems, which does not cover the use of colorimetry, in particular, due to the inherent idempotency of light emitters. A hybrid approach to organizing optical computing is proposed, combining the use of high-speed optical elements for a class of idempotent decision-making problems and digital fuzzy calculations for a class of decision-making problems that require taking into account accumulating information with the same estimates. The structure of the computer network and algorithmic procedures for implementing a hybrid architecture for decision-making have been formed. The proposed approach is applicable for intelligent decision-making systems with large fuzzy data in various fields: infrastructure, technical, medical, military, etc. Further development of the proposed approach is determined in the structuring of multilayer decision-making procedures and the development of hybrid computing procedures, as well as the architecture of optical computing.