Reducing SRAM Reading Power with Column Data Segment and Weights Correlation Enhancement for CNN Processing

Han Xu, Ziwei Li, Ziru Li, Deliang Fan, Fei Qiao, Qi Wei, Li Luo, Xinjun Liu, Huazhong Yang

Research output: Contribution to journalArticlepeer-review

Abstract

Convolutional neural network (CNN) has been widely deployed in various processors for intelligent visual signal processing. However, the large amount of activations and weights in CNN causes huge power consumption on SRAM access. Data-adaptive SRAM design is a widely studied method to reduce SRAM reading power based on the utilization of data patterns, while current designs only exploit data patterns in a coarse granularity, and have no advantages when faced with randomly distributed weight data. In this article, we propose a hardware-software co-design scheme to reduce SRAM reading power for CNN processing. First, we propose a reconfigurable data-adaptive SRAM architecture with column data segmentation (CDS-RSRAM) to utilize data patterns. Data in one column is partitioned into several segments, and finer-grained data patterns are exploited within each segment for further reading power reduction. Then, a novel training method - minimum segmented neighbor difference (miniSND) - is proposed for enhancing the correlation of weights. MiniSND improves the similarity of weights without classification accuracy degradation, thus weights could benefit from CDS-RSRAM and be read out with less power consumption. Simulation results demonstrate that the co-design scheme saves up to 66%(8b)/89%(2b) power consumption compared with 8T SRAM.

Original languageEnglish (US)
Pages (from-to)2237-2250
Number of pages14
JournalIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Volume40
Issue number11
DOIs
StatePublished - Nov 1 2021

Keywords

  • Convolutional neural network (CNN) processor
  • SRAM
  • correlation enhancement
  • data patterns
  • low power

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Reducing SRAM Reading Power with Column Data Segment and Weights Correlation Enhancement for CNN Processing'. Together they form a unique fingerprint.

Cite this