P2LSG: Powers-of-2 Low-Discrepancy Sequence Generator for Stochastic Computing

TL;DR

This paper introduces P2LSG, a novel low-discrepancy sequence generator based on Powers-of-2 VDC sequences, enhancing accuracy and energy efficiency in stochastic computing for image and video processing tasks.

Contribution

It proposes a new low-discrepancy sequence generator for stochastic computing, demonstrating improved accuracy and energy efficiency over existing methods.

Findings

Higher accuracy in image scaling and scene merging tasks.

Lower hardware cost and energy consumption.

First SC design for scene merging.

Abstract

Stochastic Computing (SC) is an unconventional computing paradigm processing data in the form of random bit-streams. The accuracy and energy efficiency of SC systems highly depend on the stochastic number generator (SNG) unit that converts the data from conventional binary to stochastic bit-streams. Recent work has shown significant improvement in the efficiency of SC systems by employing low-discrepancy (LD) sequences such as Sobol and Halton sequences in the SNG unit. Still, the usage of many well-known random sequences for SC remains unexplored. This work studies some new random sequences for potential application in SC. Our design space exploration proposes a promising random number generator for accurate and energy-efficient SC. We propose P2LSG, a low-cost and energy-efficient Low-discrepancy Sequence Generator derived from Powers-of-2 VDC (Van der Corput) sequences. We evaluate the performance of our novel bit-stream generator for two SC image and video processing case studies: image scaling and scene merging. For the scene merging task, we propose a novel SC design for the first time. Our experimental results show higher accuracy and lower hardware cost and energy consumption compared to the state-of-the-art.