A VO₂ Neuristor Based on Microstrip Line Coupling

Area of Science:

• Neuromorphic Engineering
• Materials Science
• Computational Neuroscience

Background:

• Neuromorphic networks offer superior energy efficiency compared to traditional von Neumann architectures.
• Memristors, particularly VO2 memristors, show promise for emulating biological synapses and neurons.
• Developing simple yet effective artificial neuron circuits remains a key challenge in neuromorphic computing.

Purpose of the Study:

• To investigate the simulation of neuron function using simple circuits.
• To explore the performance of VO2 memristor units coupled with microstrip lines for neuron modeling.
• To analyze the impact of distribution parameters on circuit behavior under high-frequency signals.

Main Methods:

• Two Mott VO2 memristor units were connected and coupled with microstrip lines.
• The circuit was designed to simulate the Hodgkin-Huxley neuron model.
• The influence of distribution parameters on circuit performance was analyzed under high-frequency and high-speed signals.

Main Results:

• The proposed memristor neuron circuit successfully simulated neuron action potentials.
• Observed characteristics include amplification and threshold behavior, crucial for neuronal function.
• The study highlights the viability of VO2 memristors for creating artificial neurons.

Conclusions:

• Memristor-based circuits, specifically using VO2 memristors and microstrip lines, can effectively emulate neuron action potentials.
• This approach offers a promising direction for building energy-efficient hardware-based artificial neural networks.
• Further research into circuit design and parameter optimization can enhance neuromorphic computing capabilities.