Abstract
Neuromorphic computing systems have drawn extensive attention due to their massive parallelism, the combination of storage and computation, high energy efficiency, and inherent tolerance to fault and variation. Spintronic devices, which exploit both the charge and spin properties of electrons for information processing and storage, are considered as one of the most promising candidates owing to their low power, nearly infinite endurance, good scalability, and compatibility with the complementary metal oxide semiconductor (CMOS) technology. In this chapter, the development of spintronic devices, such as magnetic tunnel junction (MTJ), spintronic memristor, skyrmion, and spin torque oscillator (STO), that mimic the functionalities of the biological synapses and neurons for neuromorphic computing is reviewed, and the challenges that need to be addressed in future research are discussed.
| Original language | English |
|---|---|
| Title of host publication | Neuromorphic Devices for Brain-Inspired Computing |
| Subtitle of host publication | Artificial Intelligence, Perception and Robotics |
| Publisher | wiley |
| Pages | 47-89 |
| Number of pages | 43 |
| ISBN (Electronic) | 9783527835317 |
| ISBN (Print) | 9783527349791 |
| DOIs | |
| State | Published - 1 Jan 2021 |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 7 Affordable and Clean Energy
Keywords
- Magnetic tunneling junction (MTJ)
- Neuromorphic computing
- Neuron
- Skyrmion
- Spin torque oscillator (STO)
- Spintronic device
- Spintronic memristor
- Synapse
Fingerprint
Dive into the research topics of 'Spintronic Neuromorphic Devices'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver