Forecasting non-linear dynamics in optical fibers with Recurrent Neural Network - GRU. My masters thesis in IIT Hyderabad.
Data Used - https://drive.google.com/file/d/11vcknmuK1t6Lawr2THt2Td0M4GZUdY60/view?usp=sharing
My work is inspired by the groundbreaking paper "Predicting ultrafast nonlinear dynamics in fibre optics with a recurrent neural network" (https://www.nature.com/articles/s42256-021-00297-z#citeas). In this paper, the authors introduced an innovative approach that uses a recurrent neural network to predict the nonlinear dynamics of optical pulses in fiber optics. I aim to extend this work by implementing a similar approach using a GRU architecture and comparing it with the results obtained in the aforementioned paper. GRUs have been shown to be more effective in handling long-term dependencies and are less computationally expensive than LSTMs, which were used in the original paper. By comparing the results obtained with the two architectures, I hope to gain new insights into the performance of different RNN models in predicting nonlinear dynamics in optical fibers.
