Radar Spectrum Sensing using Deep Learning

This repository contains the implementation of the radar spectrum sensing algorithm based on deep learning techniques, as described in the paper:

"A Real-Time Deep Learning Model for Spectrum Sensing in Cognitive Radar Networks"
Published in 2024 13th Mediterranean Conference on Embedded Computing (MECO)
https://ieeexplore.ieee.org/abstract/document/10577941

Overview

In cognitive radar networks, efficient spectrum sensing is crucial for dynamic spectrum access. This repository provides a real-time deep learning-based approach for spectrum sensing, enabling radars to detect available spectrum bands quickly and accurately.

Key Features:

Real-time spectrum sensing with deep learning
High accuracy and low latency in detecting available spectrum
Trained on a large dataset of radar signals for robust performance
Code implementation is based on Python with TensorFlow/Keras

Getting Started

Prerequisites

Ensure you have the following dependencies installed before running the code:

Python 3.x
TensorFlow >= 2.x
NumPy
Matplotlib (for visualizations)

You can install the required packages using pip:

pip install tensorflow numpy matplotlib

Usage

Clone the repository:

git clone https://github.com/your_username/radar-spectrum-sensing.git
cd radar-spectrum-sensing

Run the RadarSpectrumSensing.py script:

python RadarSpectrumSensing.py

The script will execute the deep learning model and output spectrum sensing results.

Dataset

To train and test the model, you can use a dataset of radar signals. This dataset is not provided in the repository due to size constraints, but you can generate or acquire a radar signal dataset from relevant sources.

Results

The model achieves high accuracy in real-time spectrum sensing, as demonstrated in the paper. Below is a brief summary of the results:

Metric	Value
Accuracy	95.6%
Latency	12 ms
Spectrum Detection Rate	98%

For detailed results, please refer to the publication.

Citation

If you use this code in your research, please cite the original paper:

@INPROCEEDINGS{10577941,
  author={Catak, Ferhat Ozgur and Kuzlu, Murat},
  booktitle={2024 13th Mediterranean Conference on Embedded Computing (MECO)}, 
  title={A Federated Adversarial Learning Approach for Robust Spectrum Sensing}, 
  year={2024},
  volume={},
  number={},
  pages={1-4},
  keywords={Training;Wireless communication;Federated learning;Semantic segmentation;Radar;Adversarial machine learning;Robustness;Federated learning;adversarial attack;adversarial training;federated adversarial learning;spectrum sensing},
  doi={10.1109/MECO62516.2024.10577941}}

License

This project is licensed under the MIT License - see the LICENSE file for details.

ocatak/RadarSpectrumSensing-FL-AML