NeuroComm-RF

Neuromorphic Wireless Split Computing with Resonate-and-Fire Neurons

This repository implements Neuromorphic Wireless Split Computing with Resonate-and-Fire Neurons. The code is modfied from SNNCutoff and neurocomm-msnn.

---

Getting Started

1. Install PyTorch and other dependencies:

   pip install -r requirements.txt

---

Dataset Preprocessing: SHD

1. Download SHD dataset:

   python snncutoff/preprocessing/SHD.py

---

Training

Noiseless Training

Run the following script for noiseless training:

python training.py --config examples/shd_fc.yaml

Wireless Evaluation

Run the following script for noiseless training:

python evaluation.py --config outputs/20250520_161607_819d5a0c86344666acec95d20eb43463/.configs/config.yaml

Note: you have to update the path of model in the .configs/config.yaml.

evaluation: model_path: path_to_your_model

---

Citation

For more details, please refer to the paper.

@article{wu2025neuromorphic,
  title={Neuromorphic Wireless Split Computing with Resonate-and-Fire Neurons},
  author={Wu, Dengyu and Chen, Jiechen and Poor, H Vincent and Rajendran, Bipin and Simeone, Osvaldo},
  journal={arXiv preprint arXiv:2506.20015},
  year={2025}
}