Countermeasures Against Adversarial Examples in Radio Signal Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911949379338240 |
|---|---|
| author | Zhang, Lu Lambotharan, Sangarapillai Zheng, Gan AsSadhan, Basil Roli, Fabio |
| author_facet | Zhang, Lu Lambotharan, Sangarapillai Zheng, Gan AsSadhan, Basil Roli, Fabio |
| contents | Deep learning algorithms have been shown to be powerful in many communication network design problems, including that in automatic modulation classification. However, they are vulnerable to carefully crafted attacks called adversarial examples. Hence, the reliance of wireless networks on deep learning algorithms poses a serious threat to the security and operation of wireless networks. In this letter, we propose for the first time a countermeasure against adversarial examples in modulation classification. Our countermeasure is based on a neural rejection technique, augmented by label smoothing and Gaussian noise injection, that allows to detect and reject adversarial examples with high accuracy. Our results demonstrate that the proposed countermeasure can protect deep-learning based modulation classification systems against adversarial examples. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_06796 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Countermeasures Against Adversarial Examples in Radio Signal Classification Zhang, Lu Lambotharan, Sangarapillai Zheng, Gan AsSadhan, Basil Roli, Fabio Artificial Intelligence Deep learning algorithms have been shown to be powerful in many communication network design problems, including that in automatic modulation classification. However, they are vulnerable to carefully crafted attacks called adversarial examples. Hence, the reliance of wireless networks on deep learning algorithms poses a serious threat to the security and operation of wireless networks. In this letter, we propose for the first time a countermeasure against adversarial examples in modulation classification. Our countermeasure is based on a neural rejection technique, augmented by label smoothing and Gaussian noise injection, that allows to detect and reject adversarial examples with high accuracy. Our results demonstrate that the proposed countermeasure can protect deep-learning based modulation classification systems against adversarial examples. |
| title | Countermeasures Against Adversarial Examples in Radio Signal Classification |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2407.06796 |