This repository contains my project focused on exploring and implementing Generalized Eigenvalue Decomposition (GEVD) and Denoising Source Separation (DSS) methods in signal processing.
Description: This repository contains my project focused on exploring and implementing Generalized Eigenvalue Decomposition (GEVD) and Denoising Source Separation (DSS) methods in signal processing.
Key Highlights:
-
GEVD Implementation: The repository includes a detailed implementation of the GEVD method, showcasing its utility in blind source separation. This method is particularly explored for scenarios with limited information about the source signals or the mixing process.
-
DSS Method: Denoising Source Separation is another highlight of this project. The repository provides a comprehensive look into DSS, demonstrating its effectiveness in noise reduction and signal clarity enhancement.
-
Comparative Analysis: One of the unique aspects of this project is the comparison between GEVD and DSS, especially under varying Signal-to-Noise Ratios (SNRs). This provides a nuanced understanding of each method's strengths and limitations.
-
Additional Techniques: The project also touches on Independent Component Analysis (ICA) and Principal Component Analysis (PCA), offering a broader perspective on signal processing techniques.
-
Interactive Notebooks and Scripts: All code is provided in Python, utilizing libraries like SciPy and NumPy. Interactive Jupyter Notebooks are included for an engaging and educational exploration of the methods.
-
Visualization Tools: The project leverages matplotlib for visualizing the results, making the data more accessible and understandable.
How to Use: The repository is structured to guide users through each method step-by-step. Users can explore the Jupyter Notebooks to get a hands-on experience of the implementation and results.
Applications: This project is a valuable resource for anyone interested in signal processing, data science, or machine learning, especially in fields requiring noise reduction and signal separation.
Contribution: Contributions to the project are welcome, whether it be suggestions for improvements, bug reporting, or additional features.