Three-dimensional memristive Morris-Lecar model with magnetic induction effects and its FPGA implementation.

Area of Science:

• Computational Neuroscience
• Biophysics
• Neuro-inspired Computing

Background:

• Neuron membrane potential generates magnetic induction flow.
• Existing models lack detailed characterization of these magnetic effects.

Purpose of the Study:

• To propose a 3D memristive Morris-Lecar (ML) neuron model.
• To characterize magnetic induction flow induced by neuron membrane potential.
• To investigate the impact of magnetic induction on neural firing activities.

Main Methods:

• Replaced the slow modulation current with a memristor induction current in the 3D ML model.
• Employed fast-slow analysis to elaborate bifurcation mechanisms.
• Conducted theoretical analysis and numerical simulations.
• Developed a Field-Programmable Gate Array (FPGA)-based hardware platform.

Main Results:

• The model exhibits spiking/bursting firings with period-adding bifurcation.
• Periodic and chaotic spiking-bursting patterns were observed.
• Homogeneous coexisting bursting patterns were found by switching memristor initial states.
• Experimental results from the FPGA platform validated numerical findings.

Conclusions:

• The 3D memristive ML model effectively characterizes magnetic induction effects on neural activity.
• The model demonstrates complex dynamics and potential for neuro-inspired hardware.
• Experimental validation confirms the model's accuracy and utility.