Volatile Memristor in Leaky Integrate-and-Fire Neurons: Circuit Simulation and Experimental Study

Abstract

In this paper, circuit implementation of a leaky integrate-and-fire neuron model with a volatile memristor was proposed and simulated in the SPICE simulation environment. We demonstrate that simple leaky integrate-and-fire (LIF) neuron models composed of: volatile memristor, membrane capacitance and neuron resistance can mimic spatial and temporal integration, firing function and signal decay. The existing leaky term originates from the recovery of the initial resistive state in the memristor in the spontaneous reset cycle, which is essential for emulating the forgetting process in all-memristive neural networks (MNNs). Furthermore, a diffusive perovskite memristor was used to validate the model where intrinsic memristors’ capacitance acts as neuron membrane capacitance. Good agreement with experimental and simulation results was observed. Volatility, as an inherent property of specific memristors, eliminates the need for usage of an additional peripheral circuit which will reinitialize device state, thus allowing the development of energy-efficient, large scale complex memristive neural networks. The presented circuit level model of LIF neurons can facilitate the design of MNNs.

A bi-functional three-terminal memristor applicable as an artificial synapse and neuron

“…[52][53][54][55] Depending on the properties of the LIF neuronal devices, additional external control circuits are generally required, including a capacitor, a comparator, and a pulse generator. 2,[56][57][58] As summarized in Table S1, † the reported memristor-based bi-functional devices generally require a capacitor for neuronal function emulation.…”

Artificial sensory system based on memristive devices

“…Extensive research has been conducted using the emerging memristor technology. [46,50,137,138] Artificial neurons must possess the fundamental characteristics that enable them to accumulate signals and generate spikes once the stimulus intensity surpasses a threshold. Therefore, the memristor needs specific characteristics such as threshold and relaxation behaviour, energy-efficient operation, and rapid response.…”

“…As a result, the threshold‐switching device returns to the HRS, repeating the operation of accumulating charges in the capacitor. [ 50 ] To emulate the behaviour of biological neurons, studies on threshold‐switching devices using Ag‐based memristors, metal–insulator transition materials, and chalcogenide materials have been reported.…”

“…This temporary threshold-switching characteristic of the volatile memristor is used to implement biological receptors and neurons, which also have the nature of threshold response. [45][46][47][48][49][50] On the contrary, non-volatile memristors can maintain the resistance change corresponding to an applied voltage, for a long time, making them excellent candidates for mimicking biological synapses. [51][52][53][54][55] This characteristic, combined with their scalability and power consumption that is lower than that of the traditional CMOS devices, allows them to effectively mimic the functions of biological receptors, neurons, and synapses.…”

Bifurcation and Entropy Analysis of a Chaotic Spike Oscillator Circuit Based on the S-Switch

“…A two-terminal element with an S-shaped I–V characteristic (S-IVC) [ 13 ] can be used as a switching element and is widely represented in electronics, for example, in silicon trigger diodes and thin-film structures based on oxides of transition metals (V, Nb, Ti and others) [ 16 ]. An S-switch with non-volatile memory, as in the Pt/TiO 2 /Pt structure [ 17 ], is classified as memristor, and has high potential for applications in neurotechnologies as an electronic component of neural circuits, including LIF models [ 18 , 19 ]. An S-switch with volatile memory can be built of various combinations of transistors [ 20 ], for example, as a complementary pair, where the base of one transistor is connected to the collector of another transistor.…”