Shape-Based Magnetic Domain Wall Drift for an Artificial Spintronic Leaky Integrate-and-Fire Neuron
- Univ. of Texas at Dallas, Richardson, TX (United States)
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Univ. of Texas, Austin, TX (United States)
- Istituto Nazionale di Ricerca Metrologica, Turin (Italy)
Spintronic devices based on domain wall (DW) motion through ferromagnetic nanowire tracks have received great interest as components of neuromorphic information processing systems. Prior proposals for spintronic artificial neurons required external stimuli to perform the leaking functionality, one of the three fundamental functions of a leaky integrate-and-fire (LIF) neuron. The use of this external magnetic field or electrical current stimulus results in either a decrease in energy efficiency or an increase in fabrication complexity. Here, we modify the shape of previously demonstrated three-terminal magnetic tunnel junction neurons to perform the leaking operation without any external stimuli. The trapezoidal structure causes a shape-based DW drift, thus intrinsically providing the leaking functionality with no hardware cost. This LIF neuron, thus, promises to advance the development of spintronic neural network crossbar arrays.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- Grant/Contract Number:
- AC04-94AL85000; NA0003525
- OSTI ID:
- 1581982
- Report Number(s):
- SAND-2019-8380J; SAND-2019-5379J; 677637
- Journal Information:
- IEEE Transactions on Electron Devices, Vol. 66, Issue 11; ISSN 0018-9383
- Publisher:
- IEEECopyright Statement
- Country of Publication:
- United States
- Language:
- English
Web of Science
Three Artificial Spintronic Leaky Integrate-and-Fire Neurons
|
journal | June 2020 |
Similar Records
Controllable Reset Behavior in Domain Wall–Magnetic Tunnel Junction Artificial Neurons for Task-Adaptable Computation
Maximized lateral inhibition in paired magnetic domain wall racetracks for neuromorphic computing