Reconfigurable Implementation of Fuzzy Inference System Using FPGA

Hardware implementation of algorithms is always considered to accelerate them. Fuzzy inference system (FIS) is applicable when the mathematical description of a system is inaccessible or system description is complicated and better to be described in linguistic terms. Current hardware implementation approaches requires some extended level of hardware knowledge. In this paper, we investigated current approaches in the field of FPGA-based Fuzzy controllers then a two-input generic FIS module is introduced as a baseline for constructing complicated tree based structure controller. This configurable module is independent of target hardware and needs no information about target architecture. Finally, implementation results show the benefit of this methodology in terms of speed, power, and silicon area.