Modeling coupled active resonators as an integrated photonic spiking neuron

Area of Science:

• Photonics
• Neuromorphic Engineering
• Integrated Optics

Background:

• Photonic spiking neural networks offer energy-efficient, high-throughput computation.
• Existing photonic circuits struggle with complex spiking neuron dynamics under integration constraints.

Purpose of the Study:

• To propose and model an integrated photonic spiking neuron.
• To leverage coupled active resonators for neuron dynamics.
• To address energy, footprint, and integration challenges.

Main Methods:

• Modeling using coupled-mode theory and laser rate equations.
• Simulating a system based on a saturable absorber and gain section.
• Analyzing excitable dynamics and passive Q-switching.

Main Results:

• Demonstrated controllable self-pulsation from 0.1 to 0.6 GSpikes/s.
• Achieved optical excitability with low-power optical perturbations (>30 µW).
• Observed bio-inspired leaky integrate-and-fire and refractory dynamics.

Conclusions:

• The proposed photonic spiking neuron exhibits essential spiking behaviors and bio-inspired characteristics.
• The design offers a pathway for realizing energy-efficient, high-performance neuromorphic photonic systems.
• Discussion includes design rules, fabrication platforms, and alternative choices for practical implementation.