Filters








487 Hits in 6.4 sec

Robust Adversarial Attacks Against DNN-Based Wireless Communication Systems [article]

Alireza Bahramali and Milad Nasr and Amir Houmansadr and Dennis Goeckel and Don Towsley
2021 arXiv   pre-print
In this work, we propose an input-agnostic, undetectable, and robust adversarial attack against DNN-based wireless communication systems in both white-box and black-box scenarios.  ...  We show that in the presence of defense mechanisms deployed by the communicating parties, our attack performs significantly better compared to existing attacks against DNN-based wireless systems.  ...  CONCLUSIONS In this paper, we propose an adversarial attack using a perturbation generator model against DNN-based wireless communication systems.  ... 
arXiv:2102.00918v1 fatcat:f67ecbtxavd4va7yhb7k3tkaha

Downlink Power Allocation in Massive MIMO via Deep Learning: Adversarial Attacks and Training [article]

B. R. Manoj, Meysam Sadeghi, Erik G. Larsson
2022 arXiv   pre-print
training as a defensive technique in adversarial settings and show that the robustness of DL-based wireless system against attacks improves significantly.  ...  We analyze the deep neural network (DNN) models performance against these attacks, where the adversarial perturbations are crafted using both the white-box and black-box attacks.  ...  To the best of our knowledge, this is the first work discussing adversarial training for creating robust DNN-based model for wireless systems.  ... 
arXiv:2206.06592v1 fatcat:o7tvs2f6cfawzfa577xvauux5e

Adversarial Machine Learning in Wireless Communications using RF Data: A Review [article]

Damilola Adesina, Chung-Chu Hsieh, Yalin E. Sagduyu, Lijun Qian
2021 arXiv   pre-print
This paper presents a comprehensive review of the latest research efforts focused on AML in wireless communications while accounting for the unique characteristics of wireless systems.  ...  Various methods of generating adversarial examples and attack mechanisms are also described.  ...  are robust to the effect of adversarial attacks in wireless communication systems.  ... 
arXiv:2012.14392v2 fatcat:4d3x2scwjvh33drc745mmc4gvy

Physical Adversarial Attacks Against End-to-End Autoencoder Communication Systems

Meysam Sadeghi, Erik G. Larsson
2019 IEEE Communications Letters  
We show that end-to-end learning of communication systems through deep neural network (DNN) autoencoders can be extremely vulnerable to physical adversarial attacks.  ...  We also show that classical coding schemes are more robust than autoencoders against both adversarial and jamming attacks. The codes are available at [1] .  ...  These findings suggest that defense mechanisms against adversarial attacks and further research on the security and robustness of deep-learning based wireless systems is a necessity.  ... 
doi:10.1109/lcomm.2019.2901469 fatcat:msjuyekvszeejdab4siqnoff2q

Physical Adversarial Attacks Against End-to-End Autoencoder Communication Systems [article]

Meysam Sadeghi, Erik G. Larsson
2019 arXiv   pre-print
We show that end-to-end learning of communication systems through deep neural network (DNN) autoencoders can be extremely vulnerable to physical adversarial attacks.  ...  We also show that classical coding schemes are more robust than autoencoders against both adversarial and jamming attacks. The codes are available at [1].  ...  These findings suggest that defense mechanisms against adversarial attacks and further research on the security and robustness of deep-learning based wireless systems is a necessity.  ... 
arXiv:1902.08391v1 fatcat:keiikr3xongatcpgh4afxprrsy

Channel-Aware Adversarial Attacks Against Deep Learning-Based Wireless Signal Classifiers [article]

Brian Kim, Yalin E. Sagduyu, Kemal Davaslioglu, Tugba Erpek, Sennur Ulukus
2021 arXiv   pre-print
This paper presents channel-aware adversarial attacks against deep learning-based wireless signal classifiers. There is a transmitter that transmits signals with different modulation types.  ...  Finally, a certified defense based on randomized smoothing that augments training data with noise is introduced to make the modulation classifier robust to adversarial perturbations.  ...  Jana, “Certified attacks against deep learning based power control in wireless commu- robustness to adversarial examples with differential privacy,” in IEEE nications,” in IEEE Global  ... 
arXiv:2005.05321v3 fatcat:bl5bgamxcrbrzm2thyr6fp56fu

Adversarial Attacks against Deep Learning Based Power Control in Wireless Communications [article]

Brian Kim and Yi Shi and Yalin E. Sagduyu and Tugba Erpek and Sennur Ulukus
2021 arXiv   pre-print
We consider adversarial machine learning based attacks on power allocation where the base station (BS) allocates its transmit power to multiple orthogonal subcarriers by using a deep neural network (DNN  ...  We show that the adversarial attacks are much more effective than the benchmark attack in terms of reducing the rate of communications.  ...  Moreover, since the wireless medium is shared and open to adversaries such as jammers, the adversarial attack poses a practical threat to the DNNs used in wireless communications.  ... 
arXiv:2109.08139v2 fatcat:mfdy3vqzebefnosfkgs4qhp7qa

Mixture GAN For Modulation Classification Resiliency Against Adversarial Attacks [article]

Eyad Shtaiwi, Ahmed El Ouadrhiri, Majid Moradikia, Salma Sultana, Ahmed Abdelhadi, Zhu Han
2022 arXiv   pre-print
In this paper, we propose a novel generative adversarial network (GAN)-based countermeasure approach to safeguard the DNN-based AMC systems against adversarial attack examples.  ...  However, the adversarial attacks cause the loss of accuracy for the DNN-based AMC by injecting a well-designed perturbation to the wireless channels.  ...  The current adversarial attacks on the DNN-based AMC systems adopt the white-box attack.  ... 
arXiv:2205.15743v1 fatcat:uv5mebrttrcm5btb6twt7b23fu

WiAdv

Yuxuan Zhou, Huangxun Chen, Chenyu Huang, Qian Zhang
2022 Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies  
To fill this gap, we aim to study adversarial attacks to DNN-powered WiFi-based gesture recognition to encourage proper countermeasures.  ...  the wireless medium and the inherent defects (e.g., adversarial attacks) of classifiers.  ...  Another defense method is the beamforming-based interference suppresses approach developed in wireless communication systems.  ... 
doi:10.1145/3534618 fatcat:zxydnz7cizhzpow5kvi2t6zcmu

Examining Machine Learning for 5G and Beyond through an Adversarial Lens [article]

Muhammad Usama, Rupendra Nath Mitra, Inaam Ilahi, Junaid Qadir, Mahesh K. Marina
2020 arXiv   pre-print
We also discuss approaches to mitigate this adversarial ML risk, offer guidelines for evaluating the robustness of ML models, and call attention to issues surrounding ML oriented research in 5G more generally  ...  ., resource allocation problems), there is currently tremendous excitement in the mobile networks domain around the transformative potential of data-driven AI/ML based network automation, control and analytics  ...  Towards Robust ML-Driven 5G and Beyond Networks Robustness against adversarial ML attacks is a very challenging problem.  ... 
arXiv:2009.02473v1 fatcat:ihhhcrmzb5b6bfso2wrken7owq

The Adversarial Security Mitigations of mmWave Beamforming Prediction Models using Defensive Distillation and Adversarial Retraining [article]

Murat Kuzlu, Ferhat Ozgur Catak, Umit Cali, Evren Catak, Ozgur Guler
2022 arXiv   pre-print
Experimental results show that the proposed methods effectively defend the DNN models against adversarial attacks in next-generation wireless networks.  ...  This study also offers two mitigation methods, such as adversarial training and defensive distillation, for adversarial attacks against artificial intelligence (AI)-based models used in the millimeter-wave  ...  *BS: Base Station, AP: Access Point 4 System Overview Complex Numbers and Wireless Communication The number system we use in our daily life is based on a real number system.  ... 
arXiv:2202.08185v1 fatcat:dnhf6qxajbedlon6juph73czme

Universal Adversarial Attacks on Neural Networks for Power Allocation in a Massive MIMO System [article]

Pablo Millán Santos, B. R. Manoj, Meysam Sadeghi, Erik G. Larsson
2021 arXiv   pre-print
Deep learning (DL) architectures have been successfully used in many applications including wireless systems. However, they have been shown to be susceptible to adversarial attacks.  ...  We analyze DL-based models for a regression problem in the context of downlink power allocation in massive multiple-input-multiple-output systems and propose universal adversarial perturbation (UAP)-crafting  ...  Given this security issue of DNNs, the study of adversarial attacks for wireless systems has received considerable attention [1] , [2] , [5] .  ... 
arXiv:2110.04731v1 fatcat:4x7dtet43jgfboyzsd6ozhbcce

A Survey on Adversarial Attack in the Age of Artificial Intelligence

Zixiao Kong, Jingfeng Xue, Yong Wang, Lu Huang, Zequn Niu, Feng Li, Weizhi Meng
2021 Wireless Communications and Mobile Computing  
Firstly, we explain the significance of adversarial attack. Then, we introduce the concepts, types, and hazards of adversarial attack.  ...  At the same time, adversarial attacks in the AI field are also frequent. Therefore, the research into adversarial attack security is extremely urgent.  ...  Existing defense methods are ineffective against this attack, so new defense mechanisms are needed to defend against it. 9 Wireless Communications and Mobile Computing Table 4 : 4 Continued.  ... 
doi:10.1155/2021/4907754 fatcat:rm6xcf6ryrh6ngro4sl5ifprgy

SafeAMC: Adversarial training for robust modulation recognition models [article]

Javier Maroto, Gérôme Bovet, Pascal Frossard
2021 arXiv   pre-print
In communication systems, there are many tasks, like modulation recognition, which rely on Deep Neural Networks (DNNs) models.  ...  We propose to use adversarial training, which consists of fine-tuning the model with adversarial perturbations, to increase the robustness of automatic modulation recognition (AMC) models.  ...  increases the robustness of these models; • We design a new framework based on the specific properties of communication systems.  ... 
arXiv:2105.13746v1 fatcat:vki56gfamvbpxcg2r5iz6wjvme

The RFML Ecosystem: A Look at the Unique Challenges of Applying Deep Learning to Radio Frequency Applications [article]

Lauren J. Wong, William H. Clark IV, Bryse Flowers, R. Michael Buehrer, Alan J. Michaels, William C. Headley
2020 arXiv   pre-print
in real-world wireless communication applications.  ...  communications.  ...  Two metrics are proposed in [174] for detecting adversarial attacks on wireless communications.  ... 
arXiv:2010.00432v1 fatcat:mxnvorh5wrfwzmxg4ezpbj4xve
« Previous Showing results 1 — 15 out of 487 results