A Vertical Single Transistor Neuron with Core-Shell Dual-Gate for Excitatory-Inhibitory Function and Tunable Firing...

Area of Science:

• Neuromorphic Engineering
• Solid-State Electronics
• Artificial Intelligence Hardware

Background:

• Developing efficient artificial neuron devices is crucial for advancing neuromorphic computing.
• Existing neuron models often lack the flexibility for complex learning rules and error compensation.

Purpose of the Study:

• To realize and verify a novel, multi-functional vertical nanowire single transistor neuron device.
• To demonstrate the device's capability for tunable firing threshold voltage and inhibition.
• To validate system-level performance for applications like pattern recognition.

Main Methods:

• Device fabrication and characterization using Technology Computer-Aided Design (TCAD) simulations.
• Independent control of core and shell gates for distinct functionalities.
• System-level validation through SPICE simulations of a letter recognition task.

Main Results:

• Successful realization of a core-shell dual-gate (CSDG) nanowire neuron.
• Demonstrated tunable firing threshold voltage and inhibition via independent gate control.
• Achieved myelination function for transmission rate control and successful letter recognition.

Conclusions:

• The CSDG NW neuron device offers multi-functional capabilities for advanced SNN hardware.
• This technology contributes to constructing high-density monolithic SNNs.
• The device's adaptability supports robust learning and error correction in AI systems.