Low-Power (1T1N) Skyrmionic Synapses for Spiking Neuromorphic Systems

Abstract

In this paper, we propose a “1-transistor 1-nanotrack” (1T1N) synapse based on the movement of magnetic skyrmions using spin-polarized current pulses. The proposed synaptic bit-cell has four terminals and fully decoupled spike transmission and programming paths. With careful tuning of programming parameters, we ensure multi-level non-volatile conductance evolution in the proposed skyrmionic synapse. Through micromagnetic simulations, we studied in detail the impact of the programming conditions (current density and pulse width) on synaptic performance parameters, such as the number of conductance levels and energy per transition. The programming parameters used for all further analysis gave rise to a synapse with 7 distinct conductance states and 1.2 fJ per conductance state transition event. Exploiting bidirectional conductance modulation, the 1T1N synapse is able to undergo long-term potentiation and depression according to a simplified variant of the biological spike timing-dependent plasticity rule. We present a subthreshold CMOS spike generator circuit which when coupled with a well-known subthreshold integrator circuit produces custom pre- and post-neuronal spike shapes responsible for implementing unsupervised learning with the proposed 1T1N synaptic bit-cell and consuming ~0.25 pJ/event. A spiking neural network incorporating the characteristics of the 1T1N synapse was simulated for two separate applications: pattern extraction from noisy video streams and handwritten digit classification.

Keywords

• Skyrmion
• SNN
• STDP
• neuromorphic hardware
• synapse