A high-performance and ultra-efficient fully programmable fuzzy membership function generator using FinFET technology for image enhancement

Abstract

In fuzzy systems, efficient programmable membership function generators (MFGs) are the canonical point for the fuzzification process. This work demonstrates a high-performance programmable MFG using 7 nm FinFET technology. The proposed MFG can generate S-shaped, Z-shaped, Gaussian-shaped, and Generalized Bell-shaped membership functions. The proposed design employs 14 FinFETs to control the produced waveforms’ position, height, width, and slope. According to the simulations, the proposed MFG offers remarkable improvements in transistor count (39%), power-delay product (PDP) (54%), absolute error (45%), and root mean square error (36%) compared to the previous MFGs. The proposed design has been utilized for image enhancement to evaluate the performance of the proposed MFG in realistic environments. The image enhancement simulation results indicate a higher peak signal-to-noise ratio (PSNR) and structural similarity index metric (SSIM) than previous related works. A figure of merit (FoM) is defined considering the image enhancement quality metrics and circuit efficiency to benchmark the entire performance of the proposed MFG. The FoM simulations demonstrate that the proposed design shows an excellent trade-off between the circuit performance and image enhancement quality. Our results confirm that the proposed MFG is suitable for developing high-performance on-chip image processing applications.