Abstract:
The fuzzy type-2 fuzzy inference system presents a superior performance in relation to
other methods for applications in systems with uncertainty. This work presents the architecture
of a SIF type-2 that aims to use the trapezoidal membership function that
ensures faster and simpler circuits. The architecture implemented in FPGA consists of
five blocks, fuzzifier, inference, rule base, type-reductor and defuzzifier. The hardware
presented consists of two 8-bit inputs with three trapezoidal membership functions for
each input, nine rules and an 8-bit output with three membership functions. The Verilog
implementation results are compared with the same architecture implemented in Matlab®
using the Toolbox for type-2 fuzzy.