https://github.com/alphansahin/wireless-federated-learning-with-non-coherent-over-the-air-computation

This respository consists of the source codes that allow one to realize over-the-air computation for federated edge learning by using Adalm Pluto SDRs.
https://github.com/alphansahin/wireless-federated-learning-with-non-coherent-over-the-air-computation

adalm communications federated federated-learning learning machine-learning over-the-air-computation pluto sdr signsgd wireless

Last synced: 2 days ago
JSON representation

This respository consists of the source codes that allow one to realize over-the-air computation for federated edge learning by using Adalm Pluto SDRs.

• Host: GitHub
• URL: https://github.com/alphansahin/wireless-federated-learning-with-non-coherent-over-the-air-computation
• Owner: alphansahin
• Created: 2022-12-17T16:20:21.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-04-26T14:27:05.000Z (8 months ago)
• Last Synced: 2024-11-08T14:05:17.394Z (about 2 months ago)
• Topics: adalm, communications, federated, federated-learning, learning, machine-learning, over-the-air-computation, pluto, sdr, signsgd, wireless
• Language: VHDL
• Homepage:
• Size: 22.9 MB
• Stars: 14
• Watchers: 1
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# FEELwithSDRs

This repository consists of the source codes that allow one to realize over-the-air computation for federated edge learning by using Adalm Pluto SDRs. The details are given the following paper.

The compiled pluto.frm is also available above.

# A Demonstration of Over-the-Air Computation for Federated Edge Learning - IEEE GLOBECOM 2022 Workshops

Abstract:

In this study, we propose a general-purpose synchronization method that allows a set of software-defined radios (SDRs) to transmit or receive any in-phase/quadrature data with precise timings while maintaining the baseband processing in the corresponding companion computers. The proposed method relies on the detection of a synchronization waveform in both receive and transmit directions and controlling the direct memory access blocks jointly with the processing system. By implementing this synchronization method on a set of lowcost SDRs, we demonstrate the performance of frequency-shift keying (FSK)-based majority vote (MV), i.e., an over-the-air-computation scheme for federated edge learning, and introduce the corresponding procedures. Our experiment shows that the test accuracy can reach more than 95% for homogeneous and heterogeneous data distributions without using channel state information at the edge devices.

Paper: https://arxiv.org/abs/2209.09954, https://ieeexplore.ieee.org/document/10008778