A Hybrid Training-Time and Run-Time Defense Against Adversarial Attacks in Modulation Classification-Reference-Cited by-同舟云学术

A Hybrid Training-Time and Run-Time Defense Against Adversarial Attacks in Modulation Classification

Author:

Zhang Lu¹^ORCID,Lambotharan Sangarapillai²^ORCID,Zheng Gan²^ORCID,Liao Guisheng¹^ORCID,Demontis Ambra³^ORCID,Roli Fabio⁴

Affiliation:

1. School of Electronic Engineering, Xidian University, Xi’an, China

2. Wolfson School of Mechanical, Electrical and Manufacturing Engineering, Loughborough University, Loughborough, U.K.

3. Department of Electrical and Electronic Engineering, University of Cagliari, Cagliari, Italy

4. Department of Informatics, Bioengineering, Robotics, and Systems Engineering, University of Genova, Genoa, Italy

Funder

Engineering and Physical Sciences Research Council

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

Electrical and Electronic Engineering,Control and Systems Engineering

Link

Reference15 articles.

2. CAT: Customized adversarial training for improved robustness;cheng;arXiv 2002 06789,2020

3. Towards deep learning models resistant to adversarial attacks;madry;Proc Int Conf Learn Represent,2018

4. MMA training: Direct input space margin maximization through adversarial training;ding;Proc Int Conf Learn Represent,2019

Cited by 10 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Towards Intelligent Security Mechanisms for Connected Things;Anais do XXXVII Concurso de Teses e Dissertações (CTD 2024);2024-07-21

4. Data-Driven Subsampling in the Presence of an Adversarial Actor;2024 IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN);2024-05-05