In this paper, we propose a novel training scheme to improve the robustness of end-to-end autoencoder-based channel codes against universal adversarial perturbations (UAP) and jamming attacks. Our training algorithm improves robustness by augmenting clean training data with uniformly sampled perturbations from a bounded l2-ball, encouraging consistent decoder outputs within these regions. Our contributions include the design of the training algorithm, a robustness comparison with standard autoencoders, and a benchmark against a traditional BPSK system with Hamming coding and maximum likelihood decoding. Results show that our approach enhances robustness to UAP and jamming attacks.
Publisher
Allerton Conference on Communication, Control, and Computing
Series/Report Name or Number
2025 61st Allerton Conference on Communication, Control, and Computing Proceedings
ISSN
2836-4503
Type of Resource
Text
Genre of Resource
Conference Paper/Presentation
Language
eng
Handle URL
https://hdl.handle.net/2142/130326&&
Copyright and License Information
Copyright 2025 is held by Amir Ali Belbasi and Rémi A. Chou.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.