Bioinspired Photoresponsive Single Transistor Neuron for a Neuromorphic Visual System

Area of Science:

• Neuromorphic Engineering
• Artificial Intelligence
• Materials Science

Background:

• Neuromorphic computing aims to replicate brain functions using artificial systems.
• Existing artificial visual systems often require complex hardware, limiting miniaturization and cost-effectiveness.
• A key challenge is developing low-cost, light-sensitive components for artificial visual processing.

Purpose of the Study:

• To develop a novel photoresponsive neuron device for low-cost, hardware-based neuromorphic artificial visual systems.
• To engineer an artificial neuron that mimics the light-response characteristics of biological retinal neurons.
• To demonstrate the modulation of neuron firing by optical stimuli.

Main Methods:

• Fabrication of a single-transistor artificial neuron device.
• Engineering the transistor to exhibit photoresponsive properties.
• Investigating the influence of electrical and optical stimuli on neuron firing characteristics (frequency and amplitude).
• Analyzing the effect of light intensity and wavelength on device performance.

Main Results:

• Successfully developed a single-transistor photoresponsive neuron device.
• Demonstrated that light stimuli (photons) modulate neuron firing, lowering the threshold voltage via electron-hole pair generation.
• Showcased that photoresponsive properties are tunable by light intensity and wavelength, similar to retinal neurons.
• Confirmed that the device integrates neuronal function without bulky external sensors or circuits.

Conclusions:

• The developed photoresponsive neuron device offers a low-cost, miniaturized solution for neuromorphic artificial visual systems.
• This single-transistor device effectively mimics biological neuron light response, enabling direct light interaction.
• The findings pave the way for more compact and efficient artificial vision technologies.