Exploiting Uniform Spatial Distribution to Design Efficient Random Number Source for Stochastic Computing

Stochastic computing (SC) is a promising computing paradigm for error-tolerant applications due to its low hardware cost and high fault tolerance. However, the latency of SC is often a critical bottleneck. In this work, we address this challenge by proposing a novel random number source design that exploits uniform spatial distribution.

We introduce a basic architecture to generate the uniform spatial distribution and provide a detailed implementation. We also propose optimization methods to reduce both hardware cost and improve accuracy. Our experimental results demonstrate that the proposed approach can reduce 88% area with close accuracy compared to existing random number source designs for stochastic computing.

This work was presented at the 2022 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).