Skip to main navigation Skip to search Skip to main content

Noise Analysis of Readout Chain in FDSOI-based 1T-APS for In-Sensor Vector-Matrix-Multiplication

  • Y. Xiao*
  • , Z. Zhou
  • , Y. Wang*
  • , J. Li
  • , G. Yu
  • , S. Li
  • , H. Yang
  • , L. Han
  • , R. Chen
  • , X. Liu
  • , J. Kang
  • , P. Huang
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A noise analysis scheme for readout chain of Fully-Depleted Silicon-on-Insulator (FDSOI) based one transistor active pixel sensor (1T-APS) has been proposed. Both shot noise and read noise in pixel and the transfer function of column level circuit are modelled and calibrated to describe the signal-to-noise (SNR) behavior. The SNR behavior with gate length decreasing is predicted (Peak SNR 39.8 dB @ L 60nm). The noise analysis enables accuracy recovery for in-sensor VMM assisted deep learning through noise participating in hardware-aware training and shows little deviation from baseline (<0.5% for Cifar-10 @ VGG11).

Original languageEnglish
Title of host publicationIEEE Electron Devices Technology and Manufacturing Conference
Subtitle of host publicationStrengthening the Globalization in Semiconductors, EDTM 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350371529
DOIs
StatePublished - 2024
Event8th IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2024 - Bangalore, India
Duration: 3 Mar 20246 Mar 2024

Publication series

NameIEEE Electron Devices Technology and Manufacturing Conference: Strengthening the Globalization in Semiconductors, EDTM 2024

Conference

Conference8th IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2024
Country/TerritoryIndia
CityBangalore
Period3/03/246/03/24

Keywords

  • FDSOI
  • Hardware-Aware Training
  • In-sensor Computing
  • Noise
  • Readout Chain

Fingerprint

Dive into the research topics of 'Noise Analysis of Readout Chain in FDSOI-based 1T-APS for In-Sensor Vector-Matrix-Multiplication'. Together they form a unique fingerprint.

Cite this