2020年第81回応用物理学会秋季学術講演会

講演情報

一般セッション(口頭講演)

13 半導体 » 13.5 デバイス/配線/集積化技術

[11a-Z09-1~12] 13.5 デバイス/配線/集積化技術

2020年9月11日(金) 08:30 〜 11:45 Z09

太田 健介(キオクシア)、中塚 理(名大)

10:15 〜 10:30

[11a-Z09-7] 3D Integration of RRAM Array with Oxide Semiconductor FET for In-Memory Computing

〇(P)Jixuan Wu1、Fei Mo1、Takuya Saraya1、Toshiro Hiramoto1、Masaharu Kobayashi1 (1.IIS, Univ. of Tokyo)

キーワード:3D integration, RRAM array, IGZO access transistor

In-memory computing has attracted worldwide attention for deep neural network applications because of its high energy efficiency [1]. In particular, RRAM-based neural network has been extensively studied from device to system level [2-5]. 2D neural net suffers from large energy and delay in long interconnect wires. 3D neural net is a new direction enabling area-efficient, low power, and low latency computing. In this work, we propose and develop a monolithic 3D integration of RRAM array with IGZO access transistor. Then we demonstrate basic functionality of in-memory computing in the 3D neural net.