Modeling and Simulation of Ultra Low Dispersion PCFs for High Speed Optical Communication
Implementation Plan:
—————————
Step 1:Initially, we collect and load FEM simulation data from a Photonic Crystal Fiber (PCF) design dataset .
Step 2: Then, we preprocess the data using min–max normalization, log transformation for confinement loss, and data augmentation to handle geometric variability.
Step 3: Next, we perform a feature extraction process to extract PCF geometric features and optical parameters based on collected data.
Step 4: Next, we train the data using a Deep LSTM (DLSTM) model to predict dispersion and confinement loss.
Step 5: Next, we optimize the PCF design using Genetic Algorithm (GA) and Particle Swarm Optimization (PSO).
Step 6: Next, we analyze the optimized PCF results by comparing D-LSTM predictions with FEM outputs.
Step 7: Finally, we plot graphs for the following metrics :
7.1: Number of Samples vs. Mean Absolute Error (MAE)
7.2: Number of Samples vs. Mean Squared Error (MSE)
7.3: Number of Samples vs. R² Coefficient of Determination
7.4: Number of Samples vs. Frequency-Specific Accuracy (%)
7.5: Number of Samples vs. Computational Time
Software Requirements:
—————————–
1. Development Tool: Python 3.11.x or above version
2. Operating System: Windows 10 (64-bit) or above
Dataset Link:
—————–
Link: https://www.kaggle.com/datasets/ziya07/photonic-crystal-fibers
Note:
——-
1) If the proposed plan does not fully align with your requirements, please provide all necessary details—including steps, parameters, models, and expected outcomes—in advance. Kindly ensure that any missing configurations or specifications are clearly outlined in the plan before confirming.
2) If there’s no built-in solution for what the project needs, we can always turn to reference models, customize our own, different math models or write the code ourselves to fulfil the process.
3) If the plan satisfies your requirement, Please confirm with us.
4) Project based on Simulation only.
5) If you have any changes in the dataset, kindly provide it before implementation.
