/Convolutional_SNN_RadarGesture8GHz

Primary LanguagePythonMIT LicenseMIT

A 2-μJ, 12-class, 91% Accuracy Spiking Neural Network Approach For Radar Gesture Recognition

This repository contains evaluation code for radar gesture recognition with SNNs.

Link to the paper: https://ieeexplore.ieee.org/document/9537632

Example code

Please run SNN_12_class_train.py or SNN_5_class_train.py for training the convolutional SNN.

Run SNN_12_class_test.py or SNN_5_class_test.py for a quick test using saved SNN models.

How to cite

If you use this work, please cite:

A. Safa, Federico Corradi et al., "Improving the Accuracy of Spiking Neural Networks for Radar Gesture Recognition Through Preprocessing," in IEEE Transactions on Neural Networks and Learning Systems, doi: 10.1109/TNNLS.2021.3109958.

Ali Safa et al. (2021) "A 2-μJ, 12-class, 91% Accuracy Spiking Neural Network Approach For Radar Gesture Recognition"

Acknowledgement

We would like to thank Dr. Federico Corradi for providing foundational code examples.