Improved Conductance Linearity and Conductance Ratio of 1T2R Synapse Device for Neuromorphic Systems

被引：57

作者：

Moon, Kibong ^{[1
]}

Kwak, Myounghoon ^{[1
]}

Park, Jaesung ^{[1
]}

Lee, Dongwook ^{[1
]}

Hwang, Hyunsang ^{[1
]}

机构：

[1] Pohang Univ Sci & Technol, Dept Mat Sci & Engn, Pohang 37673, South Korea

来源：

IEEE ELECTRON DEVICE LETTERS | 2017年 / 38卷 / 08期

基金：

新加坡国家研究基金会;

关键词：

Analog memory; linearity; neuromorphic; PCMO; RRAM; resistive switching; synapse device; MEMORY;

D O I：

10.1109/LED.2017.2721638

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

We report on a 1-transisor/2-resistor (1T2R) synapse device with improved conductance linearity and conductance ratio under an identical pulse condition for hardware neural networks with high pattern-recognition accuracy. Utilizing an additional series-connected resistor, the conductance linearity of a synapse device was significantly improved owing to the reduced initial voltage drop on an resistive RAM (RRAM) device during depression conditions. Moreover, to maximize the conductance ratio of a synapse device, we utilized a steep subthreshold region of an MOSFET by a parallel connection of an RRAM and a transistor. A small change in voltage on the RRAM directly controlled the gate bias of the MOSFET, which causes a large change in the drain current. Compared with a conventional RRAM synapse device, the 1T2R synapse device shows an improved conductance linearity and conductance ratio (>x100). Finally, we confirmed an excellent classification accuracy by using a neural network simulation based on a multilayer perceptron.

引用

页码：1023 / 1026

页数：4

共 12 条

[1]

[Anonymous], 2015, IEEE INT EL DEV M IE, DOI [10.1109/IEDM.2015.7409718, DOI 10.1109/IEDM.2015.7409718]

[2]

[Anonymous], 2013, IEDM

[3] Nonlinear behavior of memristive devices during tuning process and its impact on STDP learning rule in memristive neural networks [J].

Bayat, Farnood Merrikh ;

Shouraki, Saeed Bagheri .

NEURAL COMPUTING & APPLICATIONS, 2015, 26 (01) :67-75

[4] Experimental Demonstration and Tolerancing of a Large-Scale Neural Network (165 000 Synapses) Using Phase-Change Memory as the Synaptic Weight Element [J].

Burr, Geoffrey W. ;

Shelby, Robert M. ;

Sidler, Severin ;

di Nolfo, Carmelo ;

Jang, Junwoo ;

Boybat, Irem ;

Shenoy, Rohit S. ;

Narayanan, Pritish ;

Virwani, Kumar ;

Giacometti, Emanuele U. ;

Kuerdi, Bulent N. ;

Hwang, Hyunsang .

IEEE TRANSACTIONS ON ELECTRON DEVICES, 2015, 62 (11) :3498-3507

[5] Optimization of Conductance Change in Pr1-xCaxMnO3-Based Synaptic Devices for Neuromorphic Systems [J].

Jang, Jun-Woo ;

Park, Sangsu ;

Burr, Geoffrey W. ;

Hwang, Hyunsang ;

Jeong, Yoon-Ha .

IEEE ELECTRON DEVICE LETTERS, 2015, 36 (05) :457-459

[6] Ferroelectric Artificial Synapses for Recognition of a Multishaded Image [J].

Kaneko, Yukihiro ;

Nishitani, Yu ;

Ueda, Michihito .

IEEE TRANSACTIONS ON ELECTRON DEVICES, 2014, 61 (08) :2827-2833

[7] Categorization of resistive switching of metal-Pr0.7Ca0.3MnO3-metal devices [J].

Liao, Z. L. ;

Wang, Z. Z. ;

Meng, Y. ;

Liu, Z. Y. ;

Gao, P. ;

Gang, J. L. ;

Zhao, H. W. ;

Liang, X. J. ;

Bai, X. D. ;

Chen, D. M. .

APPLIED PHYSICS LETTERS, 2009, 94 (25)

[8] Comparison of Off-chip Training Methods for Neuromemristive Systems [J].

Merkel, Cory ;

Kudithipudi, Dhireesha .

2015 28TH INTERNATIONAL CONFERENCE ON VLSI DESIGN (VLSID), 2015, :99-104

[9] Analog Synapse Device With 5-b MLC and Improved Data Retention for Neuromorphic System [J].

Moon, Kibong ;

Cha, Euijun ;

Park, Jaesung ;

Gi, Sanggyun ;

Chu, Myonglae ;

Baek, Kyungjoon ;

Lee, Byunggeun ;

Oh, Sang Ho ;

Hwang, Hyunsang .

IEEE ELECTRON DEVICE LETTERS, 2016, 37 (08) :1067-1070

[10] Physical aspects of low power synapses based on phase change memory devices [J].

Suri, Manan ;

Bichler, Olivier ;

Querlioz, Damien ;

Traore, Boubacar ;

Cueto, Olga ;

Perniola, Luca ;

Sousa, Veronique ;

Vuillaume, Dominique ;

Gamrat, Christian ;

DeSalvo, Barbara .

JOURNAL OF APPLIED PHYSICS, 2012, 112 (05)

← 1 2 →