O-TI-SB-TE-based synaptic biomimetic devices for use in artificial neural networks

An artificial neural network, o-ti-sb-te technology, applied in the direction of electrical components, can solve the problems of poor linearity, large difference between high and low resistance, low bit resolution, etc., and achieve the effect of excellent electrical properties

Inactive Publication Date: 2021-05-07
HANGZHOU DIANZI UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the factors that limit the performance of artificial neural networks using non-volatile memory are the poor linear change of the resistance of the synaptic device to the number of pulses, the low bit resolution, and the large or small difference between high and low resistance.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • O-TI-SB-TE-based synaptic biomimetic devices for use in artificial neural networks
  • O-TI-SB-TE-based synaptic biomimetic devices for use in artificial neural networks
  • O-TI-SB-TE-based synaptic biomimetic devices for use in artificial neural networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] Example 1: An electrode in a medium with a diameter of 190 nanometers was prepared using a standard 0.13 micron CMOS process. The storage medium layer is deposited by physical vapor deposition with a thickness of 120 nanometers, and the O layer above the electrode is prepared through glue coating, photolithography, etching and other processes. 6 (Ti 0.4 Sb 2 Te 3 ) 94 material pattern such that O above each bottom electrode 6 (Ti 0.4 Sb 2 Te 3 ) 94 The materials are electrically insulated from each other. The top electrode titanium nitride TiN was deposited by physical vapor deposition with a thickness of 20 nanometers, and the O 6 (Ti 0.4 Sb 2 Te 3 ) 94 above the TiN electrode material pattern so that each O 6 (Ti 0.4 Sb 2 Te 3 ) 94 The upper top electrode materials are electrically insulated from each other. get based on O 6 (Ti 0.4 Sb 2 Te 3 ) 94 The T-shaped synaptic bionic unit of the material; and the performance parameters of the product a...

Embodiment 2

[0053] Example 2: An electrode in a medium with a diameter of 190 nanometers was prepared using a standard 0.13 micron CMOS process. The storage medium layer is deposited by chemical vapor deposition with a thickness of 200 nanometers, and the O layer above the electrode is prepared through glue coating, photolithography, etching and other processes. 10 (TiSb 40 Te 10 ) 90 material pattern such that O above each bottom electrode 10 (TiSb 40 Te 10 ) 90 The materials are electrically insulated from each other. The top electrode titanium nitride TiN was deposited by physical vapor deposition with a thickness of 200 nm, and O 10 (TiSb 40 Te 10 ) 90 above the TiN electrode material pattern so that each O 10 (TiSb 40 Te 10 ) 90 The upper top electrode materials are electrically insulated from each other. get based on O 10 (TiSb 40 Te 10 ) 90 The T-shaped synaptic bionic unit of the material; and the performance parameters of this product are that the difference be...

Embodiment 3

[0054] Example 3: An electrode in a medium is prepared by using a standard CMOS process, with a diameter of 10 nanometers. The storage medium layer is deposited by chemical vapor deposition with a thickness of 50 nanometers, and the O layer above the electrode is prepared through glue coating, photolithography, etching and other processes. 3 (TiSb 10 Te 10 ) 97 material pattern such that O above each bottom electrode 3 (TiSb 10 Te 10 ) 97 The materials are electrically insulated from each other. The top electrode titanium nitride TiN was deposited by physical vapor deposition with a thickness of 20 nanometers, and the O 3 (TiSb 10 Te 10 ) 97 above the TiN electrode material pattern so that each O 3 (TiSb 10 Te 10 ) 97 The upper top electrode materials are electrically insulated from each other. get based on O 3 (TiSb 10 Te 10 ) 97 The T-shaped synaptic bionic unit of the material; and the performance parameters of this product are that the difference between...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an O-Ti-Sb-Te-based synaptic bionic device used in an artificial neural network. O‑Ti‑Sb‑Te material is used as a storage medium in synaptic biomimetic devices, which can realize the conversion between multi-state resistance values ​​under the action of electrical signals. Under the operation of electrical signals, the synaptic biomimetic device based on O‑Ti‑Sb‑Te series phase change materials can provide a resolution close to 8 bits in the intermediate state resistance value, and the difference in switch resistance can reach 100 times. Under the same pulse operation The linear response of the resistance to the number of pulses is realized, and it has excellent electrical properties as a synaptic bionic device.

Description

technical field [0001] The invention relates to the field of semiconductor manufacturing materials and semiconductor devices, in particular to a storage medium material and a device for manufacturing synaptic bionic devices in artificial neural networks. Background technique [0002] Artificial Neural Network (ANN) is a research hotspot in the field of artificial intelligence since the 1980s. It abstracts the human brain neuron network from the perspective of information processing, establishes a simple model, and forms different networks according to different connection methods. In engineering and academia, it is often referred to directly as a neural network or a neural network. A neural network is an operational model consisting of a large number of nodes (or neurons) connected to each other. Each node represents a specific output function called an activation function. Each connection between two nodes represents a weighted value for the signal passing through the co...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): H01L45/00
CPCH10N70/20H10N70/881
Inventor 任堃周猷赵哲昊季振国
Owner HANGZHOU DIANZI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap