Tutorial: High-speed low-power neuromorphic systems based on magnetic Josephson junctions
Учебный обзор: Высокоскоростные энергоэффективные нейроморфные системы на основе магнитных джозефсоновских переходов
2018-10-25
SCID: 54.1/f6wk64dk
Discuss with AI
magnetic Josephson junctionsneuromorphic systemssingle flux quantum circuitsspiking neural circuitssuperconducting transmission lines
Figures from the paper
Abstract (AI)
Josephson junctions and single flux quantum (SFQ) circuits form a natural neuromorphic technology with SFQ pulses and superconducting transmission lines simulating action potentials and axons. Josephson junctions consist of superconducting electrodes with nanoscale barriers that modulate the coupling of the complex superconducting order parameter across the junction. When the order parameter undergoes a 2π phase jump, the junction emits a voltage pulse with an integrated amplitude of a flux quantum ϕ0 = h/(2e) = 2.068 × 10−15 V s. The coupling across a junction can be controlled and modulated by incorporating the nanoscale magnetic structure in the barrier. The magnetic state of embedded nanoclusters can be changed by applying small current or field pulses, enabling both unsupervised and supervised learning. The advantage of this magnetic/superconducting technology is that it combines natural spiking behavior and plasticity in a single nanoscale device and is orders of magnitude faster and lower energy than other technologies. Maximum operating frequencies are above 100 GHz, while spiking and training energies are ∼10−20 J and 10−18 J, respectively. This technology can operate close to the thermal limit, which at 4 K is considerably lower energy than in a human brain. The transition from deterministic to stochastic behavior can be studied with small temperature modifications. Here, we present a tutorial on the spiking behavior of Josephson junctions; the use of the nanoscale magnetic structure to modulate the coupling across the junction; the design and operation of magnetic Josephson junctions, device models, and simulation of magnetic Josephson junction neuromorphic circuits; and potential neuromorphic architectures based on hybrid superconducting/magnetic technology.
Key Findings
1
Embedding nanoscale magnetic structures in Josephson-junction barriers enables electrical or magnetic control of coupling, supporting both unsupervised and supervised learning.
2
Josephson junctions and SFQ circuits naturally implement neuromorphic action potentials and axons through quantized voltage pulses and superconducting transmission lines.
3
Magnetic Josephson junctions combine spiking behavior and plasticity within a single nanoscale device.
4
Operation near the thermal limit enables substantially lower energy consumption than the human brain at 4 K, while temperature tuning allows deterministic-to-stochastic behavior studies.
5
The technology offers operating frequencies above 100 GHz, with approximately 10^-20 J spiking energy and 10^-18 J training energy.
Research Object
magnetic Josephson junction-based neuromorphic systems
Research Subject
high-speed, low-power spiking behavior, plasticity, magnetic coupling modulation, and neuromorphic circuit operation
Publication Details
Publication Date
2018-10-25
Journal
Publisher
ISSN
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest