A 2941-TOPS/W Charge-Domain 10T SRAM Compute-in-Memory for Ternary Neural Network

Sungsoo Cheon, Kyeongho Lee, Jongsun Park

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)

Abstract

In this paper, we present a 10T SRAM compute-in memory (CiM) macro to process the multiplication-accumulation (MAC) operations between ternary-inputs and binary-weights. In the proposed 10T SRAM bitcell, the charge-domain analog computations are employed to improve the noise tolerance of bit-line (BL) signals where the MAC results are represented in CiM. Parallel processing of 3 different analog levels for ternary input activations is also performed in the proposed single 10T bitcell. To reduce the analog-to-digital converter (ADC) bit-resolutions without sacrificing deep neural network (DNN) accuracies, a confined-slope non-uniform integration (CS-NUI) ADC is proposed, which can provide layer-wise adaptive quantization for multiple different layers with different MAC distributions. In addition, by sharing the ADC reference voltage generator in every single column of SRAM array, the ADC area is effectively reduced with improved energy efficiencies of CiM. The 256×64.10T SRAM CiM macro with the proposed charge-sharing scheme and CS-NUI ADCs has been implemented using 28nm CMOS process. The silicon measurement results show that the proposed CiM shows the accuracies of 98.66% and 88.48% with MNIST dataset on MLP, and CIFAR-10 dataset on VGGNet-7, respectively, with the energy efficiency of 2941-TOPS/W and the area efficiency of 59.584-TOPS/mm2.

Original languageEnglish
Pages (from-to)2085-2097
Number of pages13
JournalIEEE Transactions on Circuits and Systems I: Regular Papers
Volume70
Issue number5
DOIs
Publication statusPublished - 2023 May 1

Bibliographical note

Publisher Copyright:
© 2004-2012 IEEE.

Keywords

  • Compute-in-memory (CiM)
  • binary weight
  • charge-domain computation
  • integration ADC
  • ternary input

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture

Fingerprint

Dive into the research topics of 'A 2941-TOPS/W Charge-Domain 10T SRAM Compute-in-Memory for Ternary Neural Network'. Together they form a unique fingerprint.

Cite this