Streaming-CNN FPGA Architecture for Radio Modulation Classification 🚀

Table of Contents 📚

• Introduction
• Dataset
• Installation
• License
• Contact

Introduction ✨

Showcasing the demo presented at FCCM 2022 where a streaming-based CNN FPGA architecture for streaming inputs was presented.

Dataset 📊

We use the RadioML dataset to showcase the accelerator's performance and achieve an accuracy of 94%.

Installation 🛠️

Step-by-step instructions on how to set up the project.

# Clone the repository
git clone https://github.com/axdy/rfsoc_modulation_classification.git

# Navigate to the project directory
cd rfsoc_modulation_classification

# Install package
pip install ./

Usage 🚀

• Launch the notebook voila_modulation_classification.ipynb and launch in voila mode.
• Interact with ipywidgets and experiment with accelerator running live in the programmable logic.

Contact 📬

e a.maclellan at strath.ac.uk