Design of Artificial Spiking Neuron with SiO2 Memristive Synapse to Demonstrate Neuron-Level Spike Timing Dependent Plasticity

Jessie Xuhua Niu, Hasita Veluri, Yida Li, Umesh Chand, Jin Feng Leong, Evgeny Zamburg, Maheswari Sivan, Aaron Voon-Yew Thean

Abstract: In this paper, we report on the Design-Technology Co-optimization (DTCO) of an artificial spiking neuron utilizing a leaky-integrated-and-fire (LIF) circuit with a SiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> based sub-1.2V analog resistive random-access memory (ReRAM). The memristor integrated artificial neuron is capable of spike-timing-dependent plasticity (STDP), necessary for the implementation of Spiking Neural Networks (SNNs). In addition, for the first time, we designed and demonstrated a STDP circuit in conjunction with the SiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> -based ReRAM that is capable of generating forward- and back-propagation signals. This allows for self-modulation of the synaptic weights through the Long-term potentiation (LTP) and Long-term depression (LTD) effect.