Current research on fuzzy set applications has yielded a betternunderstanding of complex design problems involving nondeterministicnhuman factors. This has in turn resulted in a significant number ofnefficient software case-study oriented realizations. Concurrently,nefficient hardware implementations involving integration of a largennumber of discrete components have become feasible in light of existingntechnological achievements. Hirota and Ozawa (1989) introduced thennotion of the fuzzy JK flip-flop, considering it an essential and basicncomponent for effective processing of fuzzy information. Continuing withnthis line of reasoning, a discussion is presented of VLSI implementationnissues with emphasis on design platforms to be used with the newnstructures. Next some comparisons are performed pertaining to thenusefulness and realization of several logical connectives, followed bynan examination of fuzzy sequential system design. The combinational partnof the circuit is embodied in a single layer neural network in whichnweights are adjusted on the basis of training situations. Numericalnconsiderations highlight the performance of the design process
展开▼
