Note: All code, data and supplementary material are used for the following paper:

Title: One-Bit Aggregation for Over-the-Air Federated Learning Against Byzantine Attacks

Author: Yifan Miao, Wanli Ni, and Hui Tian

Institution: Beijing University of Posts and Telecommunications

This paper integrates orthogonal frequency division multiplexing and SignSGD with majority vote to enhance the resilience of AirFL against Byzantine attacks by performing one-bit quantized gradient compression.

In the document Supplementary Material for 'One-Bit Aggregation for Over-the-Air Federated Learning Against Byzantine Attacks", we present the detailed derivations for Theorem 1 and Theorem 2.

@article{Miao2023One,
    author = {Yifan Miao, Hui Tian, and Wanli Ni},
    title = {One-Bit Aggregation for Over-the-Air Federated Learning Against Byzantine Attacks},
    journal = {IEEE Signal Processing Letters},
    year = {2023},
    month = {June},
    note = {Under Review}
}

Dataset: MNIST

##Requirements

Python == 3.6

Torch == 1.12.1

Tensorflow ==2.10.0