Photonic reservoir computing using a self-injection locked semiconductor laser under narrowband optical feedback

Area of Science:

• Optoelectronics
• Nonlinear Dynamics
• Computational Science

Background:

• Photonic reservoir computing offers a promising avenue for high-speed computation.
• Semiconductor lasers with optical feedback are key components in such systems.
• Achieving stable and versatile self-injection locking is crucial for enhanced performance.

Purpose of the Study:

• To propose and numerically demonstrate a photonic time-delay reservoir computing (TDRC) system.
• To utilize a self-injection locked semiconductor laser with narrowband apodized fiber Bragg grating (AFBG) feedback.
• To evaluate the system's computational ability, memory capacity, and robustness.

Main Methods:

• Implementing TDRC with a semiconductor laser subjected to optical feedback from an AFBG.
• Investigating self-injection locking in both weak and strong feedback regimes.
• Benchmarking performance using time series prediction and channel equalization tasks.

Main Results:

• The AFBG successfully suppresses relaxation oscillations and enables self-injection locking in both weak and strong feedback regimes.
• TDRC demonstrated good computational performance across both feedback regimes.
• The strong feedback regime expanded the usable feedback strength range and enhanced robustness to phase variations.

Conclusions:

• The proposed TDRC system based on self-injection locking is effective for complex computations.
• The use of AFBG feedback offers advantages over conventional methods, especially in the strong feedback regime.
• This approach provides a robust and high-performance platform for photonic computing applications.