Year: 2026
Type: article
Abstract: This work proposes a hybrid stochastic number (HSN) system based on quantum-dot cellular automata (QCA) majority logic. Unlike traditional single-bit stochastic encoding, HSN uses multiple bit streams with different weights to enhance numerical accuracy while preserving stochastic computing’s inherent error tolerance. A QCA-based encoder is proposed to convert classical binary numbers into stochastic sub-streams, and two arithmetic units are designed on this basis: a majority-logic adder and a multiplier. This architecture minimises sub-stream correlations, hardware cost, and facilitates parallel processing. The proposed 2-, 4-, and 8-bit HSN adders and multipliers were functionally simulated using QCADesigner, and their energy dissipation was estimated using QCAPro. The system was also evaluated in an image denoising experiment in which zero-mean Gaussian noise was added to a grayscale image. The experimental results show that the PSNR increased from 22.20 dB to 27.87 dB and the SSIM from 0.4663 to 0.7408. The proposed QCA-based HSN design combines numerical accuracy, small size, and noise tolerance, which makes it a candidate for low-energy computing systems.
Source: Scientific Reports
Authors: Ali Newaz Bahar, Khan A. Wahid
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7234704326
