A Flexible Memristor Model With Electronic Resistive Switching Memory Behavior and Its Application in Spiking Neural...

Area of Science:

• Materials Science
• Electronics Engineering
• Computational Neuroscience

Background:

• Memristive devices offer non-volatility, high density, and low power consumption, making them suitable for advanced computing.
• Accurate memristor models are crucial for developing practical neuromorphic systems and applications.
• Flexible memristors are gaining interest for their potential in next-generation electronic devices.

Purpose of the Study:

• To present a novel flexible memristor model with electronic resistive switching memory behavior.
• To construct and verify a mathematical and SPICE circuit model for a specific memristor.
• To demonstrate the model's application in implementing a biologically-inspired spiking neural network.

Main Methods:

• Fabrication of Ag-Au / MoSe2-doped Se / Au-Ag memristors using hydrothermal synthesis and magnetron sputtering.
• Performance testing of the memristor on an electrochemical workstation.
• Construction of mathematical and SPICE circuit models, validated against experimental data.
• Circuit implementation and simulation of a spiking neural network using the developed model.

Main Results:

• Successful fabrication and characterization of a flexible Ag-Au / MoSe2-doped Se / Au-Ag memristor.
• Development of an accurate mathematical and SPICE model that reflects the memristor's behavior.
• Demonstration of the model's effectiveness in simulating a spiking neural network with biological mechanisms.

Conclusions:

• The proposed flexible memristor model accurately represents the device's resistive switching characteristics.
• The model is effective for the circuit implementation of neuromorphic systems, specifically spiking neural networks.
• This work validates the potential of flexible memristors and accurate modeling for future electronic applications.