A mode-locked fiber laser simulation platform for self-tuning algorithm optimization

Mode-locked fiber lasers based on nonlinear polarization rotation (NPR) have been widely applied due to the simple setup, high performance and rich nonlinear dynamics. However, temperature, vibration, and stress can easily disrupt the optimized mode-locked state. To address this problem, automatic mode-locked lasers using different self-tuning algorithms are proposed in recent years. However, it is relatively difficult to verify and optimize the performance of self-tuning algorithm since the use of actual laser platforms, which hinders the development of intelligent mode-locked fiber laser. In this paper, we demonstrate a simulation platform for NPR mode-locked fiber lasers by using coupled Ginzburg–Landau equation and Jones matrix, which makes the optimization of intelligent self-tuning algorithm easier. As a proof-of-principle demonstration, genetic algorithm and human-like algorithm are implemented to prove the ability of comparing different self-tuning algorithms.