Abstract

In this paper, a novel robust, compact, and flexible neural-network model for a fiber Raman amplifier (FRA) is presented. The model can be used in various applications with promising accuracy and low requirement for memory. Analytical expressions are derived in order to make the optimal pump-power configuration much easier, and the computational time is reduced dramatically in comparison with other gain-design methods in real-time pump-power adjustment. The calculated on-off gain spectrum and the noise figure using the proposed model agree well with the experimental results. The model has a potential value in simulation and pump-power dynamic control.

© 2006 IEEE

Modeling extended L-band fiber amplifiers using neural networks trained on experimental data

Hamed Rabbani, Saber Jalilpiran, Kaboko Jean-Jacques Monga, Sophie Larochelle, and Leslie A. Rusch
Opt. Express 32(9) 16182-16198 (2024)

Modeling backward-pumped Raman amplifiers

Jonathan Hu, Brian S. Marks, Qun Zhang, and Curtis R. Menyuk
J. Opt. Soc. Am. B 22(10) 2083-2090 (2005)

Optimum design of a hybrid erbium-doped fiber amplifier/fiber Raman amplifier using particle swarm optimization

Alireza Mowla and Nosrat Granpayeh
Appl. Opt. 48(5) 979-984 (2009)

Flexible optical fiber channel modeling based on a neural network module

Rui Jiang, Zhi Wang, Tao Jia, Ziling Fu, Chao Shang, and Chongqing Wu
Opt. Lett. 48(16) 4332-4335 (2023)

Efficient and robust shooting algorithm for numerical design of bidirectionally pumped Raman fiber amplifiers

Hai Ming Jiang and Kang Xie
J. Opt. Soc. Am. B 29(1) 8-14 (2012)