Lee Sang Gu

The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 5, No. 7, pp. 1791-1800, Jul. 1998
10.3745/KIPSTE.1998.5.7.1791,   PDF Download:

Abstract

In this paper, we present parallelizing techniques for two fuzzy inference methods (Mamdani and Koczy methods) and propose an effective parallel fuzzy architecture, which si suitable for fuzzy information processing. The proposed parallel fuzzy architecture is composed of FPEs, CP, memory modules, interconnection network and Min circuits. This architecture achieves a relatively high performance and it is a generalized cascadable architecture. Here, each FPEi processes only the operations of i-th antecedent and i-th consequent. Then, all of the antecedents in the condition part are executed in parallel, as well as in the consequent part. So processors can be fully utilized and it is easy to configure the design with any number of antecedents, consequents and rules. This architecture can be used in a system requiring a rapid inference time in real-time system and/or a large expert system that has many inference variable in condition part and consequent part. It is especially well suited to the MIMO (Multiple-input, Multiple-output) than the MISO (Multiple-input, Single-output) system.

Statistics
Cite this article