TorchSig: A GNU Radio Block and New Spectrogram Tools for Augmenting ML Training

  • Phil Vallance LTS
  • Erebus Oh LTS
  • Justin Mullins LTS
  • Manbir Gulati N/A
  • Jared Hoffman Applied Insight
  • Matt Carrick Peraton Labs

Abstract

TorchSig uses machine learning (ML) to detect
and classify digitized radio frequency (RF)
signals. Recent updates and improvements to
TorchSig are given, as well as novel features
for image-only spectrogram generation and training,
reducing memory and computational burdens
and making training much faster. A new
GNU Radio out-of-tree (OOT) block is provided
which uses a TorchSig ML model for detecting
signals in real-time.

Published
2024-09-24
How to Cite
VALLANCE, Phil et al. TorchSig: A GNU Radio Block and New Spectrogram Tools for Augmenting ML Training. Proceedings of the GNU Radio Conference, [S.l.], v. 9, n. 1, sep. 2024. Available at: <https://pubs.gnuradio.org/index.php/grcon/article/view/147>. Date accessed: 10 oct. 2024.