Modulation Classification using Machine Learning

This project classifies different modulation types (BPSK, QPSK, 16-QAM) using machine learning techniques.
It extracts features from SDR (Software Defined Radio) signals, trains a model, and evaluates performance.

📌 Features

• Generates simulated signals that are suitable for different modulations
• Extracts key features from signal data
• Preprocesses data for training
• Trains a machine learning model and creates an ensemble model
• Evaluates and tests model performance
• Visualizes classification results

📂 File Structure

• 1_signal_generation.py - Generates synthetic modulated signals
• 2_feature_extraction.py - Extracts signal features and builds dataset
• 3_train_test_model.py - Trains and tests ML models
• 4_real_time_15seconds.py - Real-time monitoring and modulation decision
• 5_dashboard.py - Visualizes predictions and performance
• README.md - Project documentation and instructions for setup and usage

📊 Dataset

The project works with synthetic signal data saved in CSV format:

• features.csv - Extracted features from signals
• processed_features.csv - Normalized features for training
• classified_results.csv - Model predictions

Output Structure

• output/ - Stores results and metrics
• metadata/ - Config or support files (for signal data)

🛠 Dependencies

See requirements.txt

🚀 Running the Project

See RUN_INSTRUCTIONS.md