An ultra-compact leaky integrate-and-fire neuron with long and tunable time constant utilizing pseudo resistors for spiking neural networks

Abstract

Spiking neural networks (SNNs) inspired by biological neurons enable a more realistic mimicry of the human brain. To realize SNNs similar to large-scale biological networks, neuron circuits with high area efficiency are essential. In this paper, we propose a compact leaky integrate-and-fire (LIF) neuron circuit with a long and tunable time constant, which consists of a capacitor and two pseudo resistors (PRs). The prototype chip was fabricated with TSMC 65 nm CMOS technology, and it occupies a die area of 1392 μm2. The fabricated LIF neuron has a power consumption of 6 μW and a leak time constant of up to 1.2 ms (the resistance of PR is up to 600 MΩ). In addition, the time constants are tunable by changing the bias voltage of PRs. Overall, this proposed neuron circuit facilitates the very-large-scale integration (VLSI) of adaptive SNNs, which is crucial for the implementation of bio-scale brain-inspired computing.

Artificial Neurons Using Ag−In−Zn−S/Sericin Peptide‐Based Threshold Switching Memristors for Spiking Neural Networks

“…[25] As for the action mechanism of biological neurons depicted in Figure 3a, if the neurons receive sufficient input spikes from the adjacent neurons through joint synapses, its membrane potential may exceed a certain membrane potential threshold, leading to the transmition of the spike stimuli into the axon, which is often referred to as output firing. [6] Simply put, one fundamental function of biological neurons, that is, the complete realization of information exchange, can be achieved by receiving, integrating, transmitting, and outputting information. [26] It has been widely reported that the CFs-based TS devices can be used in realizing various neuronal characteristics.…”

“…[5] Structurally speaking, a hardware friendly and energy efficient SNN is composed of two fundamental information processing elements, that is, artificial neurons and synapses, which need to meet the requirements of high-density integration and excellent stability. [6,7] In order to guarantee the synapse matrix works effectively, volatile threshold switching (TS) devices are needed to simulate integrate-and-fire function of neurons. [8] Nevertheless, many neural devices used for SNNs are the mainstream complementary metal-oxidesemiconductor (CMOS) neurons that are indispensable for realizing of sophisticated circuits, which are considered unsuitable for large-scale network implementation due to their large area and low efficiency.…”