2019 IEEE 11th International Memory Workshop (IMW) 2019
DOI: 10.1109/imw.2019.8739720
|View full text |Cite
|
Sign up to set email alerts
|

Neuromorphic computing based on Analog ReRAM as low power solution for edge application

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
12
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
4
3

Relationship

0
7

Authors

Journals

citations
Cited by 17 publications
(13 citation statements)
references
References 3 publications
0
12
0
Order By: Relevance
“…30,31 The volume increases of a-TaO x observed in the (2) V maxdependent analog reset phenomenon (+1.5 to +2.5 V in Figures 2c and S5) suggested that the resistance increases in this type of switching were produced by the electrochemical oxidation of a-TaO x (second left illustration in Figure 6b). The oxidation from air (replacement of an oxygen reservoir layer in a practical device), 5,10,28 which was suggested for I c ≥ 10 μA in the over-set, should also progress in this reset switching at approximately 150 μA and will steeply increase the resistivity of a-TaO x at the interface. The reversibility of the structural changes (Figures 2c,d and S7) and stability of the cycling characteristics (Figure 2b) indicated that the redox reactions involved in this switching are highly continuous and reversible (Figure 6b), possibly due to the low probability of phase separation in a-TaO x for 2.0 < x < 2.5.…”
Section: C-afm Observation Of Time-dependent Analogmentioning
confidence: 99%
See 2 more Smart Citations
“…30,31 The volume increases of a-TaO x observed in the (2) V maxdependent analog reset phenomenon (+1.5 to +2.5 V in Figures 2c and S5) suggested that the resistance increases in this type of switching were produced by the electrochemical oxidation of a-TaO x (second left illustration in Figure 6b). The oxidation from air (replacement of an oxygen reservoir layer in a practical device), 5,10,28 which was suggested for I c ≥ 10 μA in the over-set, should also progress in this reset switching at approximately 150 μA and will steeply increase the resistivity of a-TaO x at the interface. The reversibility of the structural changes (Figures 2c,d and S7) and stability of the cycling characteristics (Figure 2b) indicated that the redox reactions involved in this switching are highly continuous and reversible (Figure 6b), possibly due to the low probability of phase separation in a-TaO x for 2.0 < x < 2.5.…”
Section: C-afm Observation Of Time-dependent Analogmentioning
confidence: 99%
“…A variety of neuromorphic functions, including high-performance analog resistive switching, [3][4][5][6][7][8]10,13 spike timing-dependent plasticity, 4,7 and second-order memristor characteristics, 7 have been demonstrated in thin-film devices of a-TaO x , in addition to the fact that it has been widely used as a practical material for commercial resistive random access memory (ReRAM). 3,10,28 In amorphous metal oxides, three types of analog switching phenomena have been demonstrated depending on the control parameters: (1) analog set (resistance decrease) controlled by the compliance current (I c ), 1,5,6,13 (2) analog reset (resistance increase) controlled by the maximum applied voltage (V max ), 1,3,4,6,8,13 and (3) analog set and reset by multiple voltage applications (i.e., controlled by the voltage application time). 2,4,7,8 In our measurements, all three types of phenomena were directly demonstrated by C-AFM, and the involved ionic migration was observed in the angstrom scale.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…Panasonic developed neuromorphic computing based on analog RRAM, resistive analog neuromorphic device (RAND), as a low power solution for edge application [116]. The authored demonstrated MNIST recognition and sensor application in which several networks could be configured at the same time.…”
Section: Analog Deep Learning Acceleratormentioning
confidence: 99%
“…Oxide-based resistive random-access memory (ReRAM) is emerging as a next generation non-volatile memory (NVM) for various applications, such as embedded NVM [240] , storage-class memory [224], [241] , and neuromorphic computing [242] .…”
Section: Introductionmentioning
confidence: 99%