Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

CMOS-Compatible Oxide TFT-Based 2T-2C Near-Pixel Analog Compute Cell with ADC/DAC-Free Sensor Interface for Energy-Efficient Feature Extraction

Full metadata record
DC Field Value Language
dc.contributor.authorPark, Seonjae-
dc.contributor.authorKim, Seungyeob-
dc.contributor.authorCheon, Seungmin-
dc.contributor.authorJung, Taeseung-
dc.contributor.authorAhn, Jinho-
dc.contributor.authorJeon, Sanghun-
dc.date.accessioned2026-04-20T07:30:13Z-
dc.date.available2026-04-20T07:30:13Z-
dc.date.issued2026-01-
dc.identifier.issn0163-1918-
dc.identifier.issn2156-017X-
dc.identifier.urihttps://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/212256-
dc.description.abstractThis work presents a CMOS-compatible 2T-2C near-pixel compute cell based on oxide TFTs with an ADC/DAC-free sensor interface, enabling energy-efficient analog-domain feature extraction for vision systems. The cell integrates IGZO TFTs and HfO2 capacitors to support signed-weight MAC operations without requiring ADC/DAC. A comparator circuit allows direct analog-to-pulse conversion from a photodiode, enabling one-to-one sensor-to-cell mapping. The fabricated device demonstrates linear weight updates, strong retention, and calibrated MAC accuracy (R2 = 0.992). Compared to 2T-0C, it achieves up to 93% lower power and up to 43% smaller footprint in peripheral circuits. CNN simulations using 2T-2C-based feature maps achieve 99.6% classification accuracy, highlighting its potential for low-power, high-accuracy near-sensor vision.-
dc.format.extent4-
dc.language영어-
dc.language.isoENG-
dc.publisherInstitute of Electrical and Electronics Engineers Inc.-
dc.titleCMOS-Compatible Oxide TFT-Based 2T-2C Near-Pixel Analog Compute Cell with ADC/DAC-Free Sensor Interface for Energy-Efficient Feature Extraction-
dc.typeArticle-
dc.publisher.location미국-
dc.identifier.doi10.1109/IEDM50572.2025.11353863-
dc.identifier.scopusid2-s2.0-105033541979-
dc.identifier.wosid001701480300270-
dc.identifier.bibliographicCitation2025 IEEE International Electron Devices Meeting (IEDM), pp 1 - 4-
dc.citation.title2025 IEEE International Electron Devices Meeting (IEDM)-
dc.citation.startPage1-
dc.citation.endPage4-
dc.type.docTypeConference paper-
dc.description.isOpenAccessN-
dc.description.journalRegisteredClassscopus-
dc.relation.journalResearchAreaComputer Science-
dc.relation.journalResearchAreaEngineering-
dc.relation.journalResearchAreaPhysics-
dc.relation.journalWebOfScienceCategoryComputer Science, Hardware & Architecture-
dc.relation.journalWebOfScienceCategoryEngineering, Electrical & Electronic-
dc.relation.journalWebOfScienceCategoryPhysics, Applied-
dc.subject.keywordPlusAnalog to digital conversion-
dc.subject.keywordPlusCells-
dc.subject.keywordPlusCMOS integrated circuits-
dc.subject.keywordPlusComparator circuits-
dc.subject.keywordPlusEnergy efficiency-
dc.subject.keywordPlusExtraction-
dc.subject.keywordPlusLow power electronics-
dc.subject.keywordPlusPixels-
dc.subject.keywordPlusThin film transistors-
dc.identifier.urlhttps://ieeexplore.ieee.org/document/11353863-
Files in This Item
Go to Link
Appears in
Collections
서울 공과대학 > 서울 신소재공학부 > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Ahn, Jinho photo

Ahn, Jinho
COLLEGE OF ENGINEERING (SCHOOL OF MATERIALS SCIENCE AND ENGINEERING)
Read more

Altmetrics

Total Views & Downloads

BROWSE