TorchSig Downloads

Code, Data, Models, and Papers

Downloads

TorchSig openly shares code, datasets, pretrained models, and publications.

TorchSig Codebase

The TorchSig codebase is freely available at our GitHub page: TorchSig GitHub

Signal Datasets

The Sig53 and Wideband Sig53 signals datasets can be generated after downloading TorchSig. Note: we plan to host these dataset for downloading at a future date, but currently it is limited to local generation via TorchSig.

Pretrained Models

TorchSig's model API can automatically reach out and download our models pretrained on the Sig53 or WBSig53 datasets, and they are additionally available for download at the below locations:

Sig53 Pre-Trained Models:

WidebandSig53 Pre-Trained Models for Signal Detection:

WidebandSig53 Pre-Trained Models for Modulation Family Recognition:

Publications

  • Large Scale Radio Frequency Signal Classification

    • Abstract: Existing datasets used to train deep learning models for narrowband radio frequency (RF) signal classification lack enough diversity in signal types and channel impairments to sufficiently assess model performance in the real world. We introduce the Sig53 dataset consisting of 5 million synthetically-generated samples from 53 different signal classes and expertly chosen impairments. We also introduce TorchSig, a signals processing machine learning toolkit that can be used to generate this dataset. TorchSig incorporates data handling principles that are common to the vision domain, and it is meant to serve as an open-source foundation for future signals machine learning research. Initial experiments using the Sig53 dataset are conducted using state of the art (SoTA) convolutional neural networks (ConvNets) and Transformers. These experiments reveal Transformers outperform ConvNets without the need for additional regularization or a ConvNet teacher, which is contrary to results from the vision domain. Additional experiments demonstrate that TorchSig's domain-specific data augmentations facilitate model training, which ultimately benefits model performance. Finally, TorchSig supports on-the-fly synthetic data creation at training time, thus enabling massive scale training sessions with virtually unlimited datasets.

  • Large Scale Radio Frequency Wideband Signal Detection & Recognition

    • Abstract: The applications of deep learning to the radio frequency (RF) domain have largely concentrated on the task of narrowband signal classification after assuming the signals of interest have already been detected and extracted from a wideband capture. We introduce the WidebandSig53 (WBSig53) dataset consisting of 550 thousand synthetically-generated samples from 53 different signal classes containing approximately 2 million unique signals. We extend the TorchSig signals processing machine learning toolkit for open-source and customizable generation, augmentation, and processing of the WBSig53 dataset. Initial experiments using the WBSig53 dataset are conducted using state of the art (SoTA) convolutional neural networks and transformers. We investigate the performance of signal detection tasks, i.e. strictly detect the presence, time, and frequency of all signals present in the input data, as well as the performance of signal recognition tasks, where networks detect the presence, time, frequency, and modulation family of all signals present in the input data. We evaluate two main approaches to these tasks with segmentation networks and object detection networks operating on complex input spectrograms. Finally, we conduct comparative analysis of the various approaches in terms of the networks’ mean average precision scores and the speed of inference.