Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems

Compute‐in‐memory (CIM) technology based on emerging nonvolatile memories (NVMs) has shown promise in enhancing artificial intelligence applications by integrating computation directly within NVM arrays. However, the efficiency of CIM systems is often curtailed by the substantial overhead that is ca...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiseong Im, Jonghyun Ko, Joon Hwang, Jangsaeng Kim, Wonjun Shin, Ryun‐Han Koo, Minkyu Park, Sung‐Ho Park, Woo Young Choi, Jae‐Joon Kim, Jong‐Ho Lee
Format: Article
Language:English
Published: Wiley 2025-05-01
Series:Advanced Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1002/aisy.202400594
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850261471528747008
author Jiseong Im
Jonghyun Ko
Joon Hwang
Jangsaeng Kim
Wonjun Shin
Ryun‐Han Koo
Minkyu Park
Sung‐Ho Park
Woo Young Choi
Jae‐Joon Kim
Jong‐Ho Lee
author_facet Jiseong Im
Jonghyun Ko
Joon Hwang
Jangsaeng Kim
Wonjun Shin
Ryun‐Han Koo
Minkyu Park
Sung‐Ho Park
Woo Young Choi
Jae‐Joon Kim
Jong‐Ho Lee
author_sort Jiseong Im
collection DOAJ
description Compute‐in‐memory (CIM) technology based on emerging nonvolatile memories (NVMs) has shown promise in enhancing artificial intelligence applications by integrating computation directly within NVM arrays. However, the efficiency of CIM systems is often curtailed by the substantial overhead that is caused by traditional complementary metal‐oxide‐semiconductor (CMOS)‐based analog‐to‐digital converters (ADCs). Here, we report an in‐memory ADC (IMADC) that leverages NVMs to perform the dual functionalities of reference generation and voltage comparison, effectively minimizing the area occupancy and energy consumption, is reported. The IMADC not only significantly outperforms traditional ADCs but also enables the inherent processing of nonlinear activation functions such as the sigmoid function, which is required for neural networks. The IMADC‐based CIM system achieves software‐comparable accuracy in CIFAR‐10 image classification on the VGG‐9 network. The IMADC exhibits significantly reduced area occupancy (45 μm2) and energy consumption (29.6 fJ) compared to conventional CMOS‐based ADCs. The IMADC, compatible with various types of NVMs, demonstrates significant potential for enhancing the efficiency of CIM systems in terms of area occupancy and energy consumption.
format Article
id doaj-art-8fed5d306cd74fe285cdbab8dba83627
institution OA Journals
issn 2640-4567
language English
publishDate 2025-05-01
publisher Wiley
record_format Article
series Advanced Intelligent Systems
spelling doaj-art-8fed5d306cd74fe285cdbab8dba836272025-08-20T01:55:22ZengWileyAdvanced Intelligent Systems2640-45672025-05-0175n/an/a10.1002/aisy.202400594Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory SystemsJiseong Im0Jonghyun Ko1Joon Hwang2Jangsaeng Kim3Wonjun Shin4Ryun‐Han Koo5Minkyu Park6Sung‐Ho Park7Woo Young Choi8Jae‐Joon Kim9Jong‐Ho Lee10Department of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Semiconductor Convergence Engineering Sungkyunkwan University Suwon 16419 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaDepartment of Electrical and Computer Engineering Inter‐University Semiconductor Research Center College of Engineering Seoul National University Seoul 08826 Republic of KoreaCompute‐in‐memory (CIM) technology based on emerging nonvolatile memories (NVMs) has shown promise in enhancing artificial intelligence applications by integrating computation directly within NVM arrays. However, the efficiency of CIM systems is often curtailed by the substantial overhead that is caused by traditional complementary metal‐oxide‐semiconductor (CMOS)‐based analog‐to‐digital converters (ADCs). Here, we report an in‐memory ADC (IMADC) that leverages NVMs to perform the dual functionalities of reference generation and voltage comparison, effectively minimizing the area occupancy and energy consumption, is reported. The IMADC not only significantly outperforms traditional ADCs but also enables the inherent processing of nonlinear activation functions such as the sigmoid function, which is required for neural networks. The IMADC‐based CIM system achieves software‐comparable accuracy in CIFAR‐10 image classification on the VGG‐9 network. The IMADC exhibits significantly reduced area occupancy (45 μm2) and energy consumption (29.6 fJ) compared to conventional CMOS‐based ADCs. The IMADC, compatible with various types of NVMs, demonstrates significant potential for enhancing the efficiency of CIM systems in terms of area occupancy and energy consumption.https://doi.org/10.1002/aisy.202400594analog‐to‐digital convertercompute‐in‐memoryflash thin‐film‐transistorhardware‐based artificial intelligenceneuromorphic computing
spellingShingle Jiseong Im
Jonghyun Ko
Joon Hwang
Jangsaeng Kim
Wonjun Shin
Ryun‐Han Koo
Minkyu Park
Sung‐Ho Park
Woo Young Choi
Jae‐Joon Kim
Jong‐Ho Lee
Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems
Advanced Intelligent Systems
analog‐to‐digital converter
compute‐in‐memory
flash thin‐film‐transistor
hardware‐based artificial intelligence
neuromorphic computing
title Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems
title_full Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems
title_fullStr Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems
title_full_unstemmed Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems
title_short Multifunctional In‐Memory Analog‐to‐Digital Converter for Next‐Gen Compute‐in‐Memory Systems
title_sort multifunctional in memory analog to digital converter for next gen compute in memory systems
topic analog‐to‐digital converter
compute‐in‐memory
flash thin‐film‐transistor
hardware‐based artificial intelligence
neuromorphic computing
url https://doi.org/10.1002/aisy.202400594
work_keys_str_mv AT jiseongim multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT jonghyunko multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT joonhwang multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT jangsaengkim multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT wonjunshin multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT ryunhankoo multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT minkyupark multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT sunghopark multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT wooyoungchoi multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT jaejoonkim multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems
AT jongholee multifunctionalinmemoryanalogtodigitalconverterfornextgencomputeinmemorysystems