Evaluating Adversarial Evasion Attacks in the Context of Wireless Communications
Open accessThe DOI matches its Crossref record; 32 of the 32 references it cites were found in a registry or an index.
Abstract
Recent advancements in radio frequency machine learning (RFML) have demonstrated the use of raw in-phase and quadrature (IQ) samples for multiple spectrum sensing tasks. Yet, deep learning techniques have been shown, in other applications, to be vulnerable to adversarial machine learning (ML) techniques, which seek to craft small perturbations that are added to the input to cause a misclassification. The current work differentiates the threats that adversarial ML poses to RFML systems based on where the attack is executed from: direct access to classifier input, synchronously transmitted over the air (OTA), or asynchronously transmitted from a separate device. Additionally, the current work develops a methodology for evaluating adversarial success in the context of wireless communications, where the primary metric of interest is bit error rate and not human perception, as is the case in image recognition. The methodology is demonstrated using the well known Fast Gradient Sign Method to evaluate the vulnerabilities of raw IQ based Automatic Modulation Classification and concludes RFML is vulnerable to adversarial examples, even in OTA attacks. However, RFML domain specific receiver effects, which would be encountered in an OTA attack, can present significant impairments to adversarial evasion.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
The Limitations of Deep Learning in Adversarial Settings
Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z. B., & Swami, A. (2016) · 2016 IEEE European Symposium on Security and Privacy (EuroS&P) · doi:10.1109/eurosp.2016.36 · Cited by 3,993
Practical Black-Box Attacks against Machine Learning
Papernot, N., McDaniel, P., Goodfellow, I., Jha, S., Celik, Z. B., & Swami, A. (2017) · Proceedings of the 2017 ACM on Asia Conference on Computer and Communications Security · doi:10.1145/3052973.3053009 · Cited by 3,495 · cited 1×
“In a very limited case, the attacker may only have access to what is deemed an oracle, an entity that will label a limited number of X,Y pairs for the attacker through an API [31] or an observable wireless transmission [16], [28].”
BackgroundDeep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images
Nguyen, A., Yosinski, J., & Clune, J. (2014) · arXiv · doi:10.48550/arxiv.1412.1897 · Cited by 3,258
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
Nguyen, A., Yosinski, J., & Clune, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298640 · Cited by 3,258
An Introduction to Deep Learning for the Physical Layer
O'Shea, T., & Hoydis, J. (2017) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2017.2758370 · Cited by 2,980
Boosting Adversarial Attacks with Momentum
Dong, Y., Liao, F., Pang, T., Su, H., Zhu, J., Hu, X., & Li, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00957 · Cited by 2,956 · cited 2×
“Targeted misclassification are sometimes split between attacks that start with a real input [19], [25] versus those that start with noise [30].”
BackgroundUniversal Adversarial Perturbations
Moosavi-Dezfooli, S.-M., Fawzi, A., Fawzi, O., & Frossard, P. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.17 · Cited by 2,680
Ensemble Adversarial Training: Attacks and Defenses
Tramèr, F., Kurakin, A., Papernot, N., Goodfellow, I., Boneh, D., & McDaniel, P. (2017) · arXiv · doi:10.48550/arxiv.1705.07204 · Cited by 2,176 · cited 1×
“Ensemble adversarial training [36] has been shown as an effective method for hardening DNN models in the CV domain and the results presented in the current work on BER penalties for adversarial examples can be used to guide”
MethodConvolutional Radio Modulation Recognition Networks
O’Shea, T. J., Corgan, J., & Clancy, T. C. (2016) · Communications in Computer and Information Science · doi:10.1007/978-3-319-44188-7_16 · Cited by 1,584 · cited 6×
“The authors present a study of evasion attacks against a similar neural network architecture [8] using the RML2016.”
BackgroundMethodResultInfluentialSurvey of automatic modulation classification techniques: classical approaches and new trends
Dobre, O. A., Abdi, A., Bar-Ness, Y., & Su, W. (2007) · IET Communications · doi:10.1049/iet-com:20050176 · Cited by 1,393 · cited 2×
“Prior approaches to spectrum sensing were likelihood or feature based [2]–[7] while more recent approaches leverage the advances in DNN to operate directly on raw IQ samples [8]–[13].”
MethodTrojaning Attack on Neural Networks
Liu, Y., Ma, S., Aafer, Y., Lee, W.-C., Zhai, J., Wang, W., & Zhang, X. (2018) · Proceedings 2018 Network and Distributed System Security Symposium · doi:10.14722/ndss.2018.23291 · Cited by 1,270 · cited 1×
“Causative attacks exert influence over the training process to inject vulnerabilities into the trained classifier [21], [22].”
BackgroundHuang, L., Joseph, A. D., Nelson, B., Rubinstein, B. I. P., & Tygar, J. D. (2011) · Proceedings of the 4th ACM workshop on Security and artificial intelligence · doi:10.1145/2046684.2046692 · Cited by 1,194 · cited 2×
“Exploratory attacks [14] seek to learn information about the classifier.”
BackgroundManipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning
Jagielski, M., Oprea, A., Biggio, B., Liu, C., Nita-Rotaru, C., & Li, B. (2018) · 2018 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2018.00057 · Cited by 861
Defense against Primary User Emulation Attacks in Cognitive Radio Networks
Chen, R., Park, J.-M. J., & Reed, J. H. (2008) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2008.080104 · Cited by 657 · cited 1×
“Adversarial machine learning could be used, in the context of RFML, to disrupt DSA systems through primary user emulation [17], evade mobile transmitter tracking [18], or avoid demodulation by confusing an Automatic Modulation Classification (AMC) system [2].”
MethodDeep architectures for modulation recognition
West, N. E., & O'Shea, T. (2017) · 2017 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN) · doi:10.1109/dyspan.2017.7920754 · Cited by 567 · cited 2×
“5) Automatic Modulation Classification Evaluation: Top-1 accuracy is the metric used for classifier evaluation in [8], [9], and [33] and is the metric we use for evaluation in the current work.”
MethodAdversarial Attacks on Deep-Learning Based Radio Signal Classification
Sadeghi, M., & Larsson, E. G. (2019) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2018.2867459 · Cited by 346
Radio Machine Learning Dataset Generation with GNU Radio
O’Shea, T. J. (2025) · TIB Data Manager · doi:10.57702/h54kawox · Cited by 303
An overview of feature-based methods for digital modulation classification
Hazza, A., Shoaib, M., Alshebeili, S. A., & Fahad, A. (2013) · International Conference on Communications, Signal Processing, and their Applications (ICCSPA) · doi:10.1109/iccspa.2013.6487244 · Cited by 239
Adversarial Transformation Networks: Learning to Generate Adversarial Examples
Baluja, S., & Fischer, I. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.09387 · Cited by 194
Modulation recognition using hierarchical deep neural networks
Karra, K., Kuzdeba, S., & Petersen, J. (2017) · IEEE International Symposium on Dynamic Spectrum Access Networks · doi:10.1109/dyspan.2017.7920746 · Cited by 116
Adversarial Deep Learning for Cognitive Radio Security: Jamming Attack and Defense Strategies
Shi, Y., Sagduyu, Y. E., Erpek, T., Davaslioglu, K., Lu, Z., & Li, J. H. (2018) · 2018 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccw.2018.8403655 · Cited by 111 · cited 3×
“In [16] and [28], the authors consider adversarial machine learning for intelligently jamming a deep learning enabled transmitter, at transmission time and sensing time, to prevent a transmission.”
BackgroundClustering Learned CNN Features from Raw I/Q Data for Emitter Identification
Wong, L. J., Headley, W. C., Andrews, S., Gerdes, R., & Michaels, A. J. (2018) · IEEE Military Communications Conference (MILCOM) · doi:10.1109/milcom.2018.8599847 · Cited by 82 · cited 2×
“Prior approaches to spectrum sensing were likelihood or feature based [2]–[7] while more recent approaches leverage the advances in DNN to operate directly on raw IQ samples [8]–[13].”
BackgroundMethodSelf-citationSpectrum Data Poisoning with Adversarial Deep Learning
Shi, Y., Erpek, T., Sagduyu, Y., & Li, J. H. (2018) · IEEE Military Communications Conference · doi:10.1109/milcom.2018.8599832 · Cited by 76 · cited 2×
“In [16] and [28], the authors consider adversarial machine learning for intelligently jamming a deep learning enabled transmitter, at transmission time and sensing time, to prevent a transmission.”
BackgroundDistributed Cyclic Spectrum Feature-Based Modulation Classification
Headley, W. C., Reed, J. D., & da Silva, C. R. C. M. (2008) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc.2008.216 · Cited by 73
Self-citationSignal detection effects on deep neural networks utilizing raw IQ for modulation classification
Hauser, S. C., Headley, W. C., & Michaels, A. J. (2017) · IEEE Military Communications Conference (MILCOM) · doi:10.1109/milcom.2017.8170853 · Cited by 55 · cited 7×
“uses extended filter sizes as done in [9] and [32], using filters with 7 taps and padded with 3 zeros on either side.”
BackgroundMethodInfluentialSelf-citationDetecting Adversarial Samples for Deep Neural Networks through Mutation TestingPreprint
Wang, J., Sun, J., Zhang, P., & Wang, X. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.05010 · Cited by 51 · cited 4×
“Furthermore, the current work has reinforced the viability of mutation testing [35] by showing that RFML”
BackgroundMethodInfluentialPractical Issues for Spectrum Management With Cognitive Radios
Dudley, S., Headley, W. C., Lichtman, M., Imana, E. Y., Ma, X., Abdelbar, M., Padaki, A. V., Ullah, A., Sohul, M. M., Yang, T., et al. (2014) · Proceedings of the IEEE · doi:10.1109/jproc.2014.2298437 · Cited by 46 · cited 1×
“of primary and secondary users in order to adapt transmission parameters to the environment [1] and has obvious applications to signals intelligence.”
BackgroundSelf-citationSecurity threats to signal classifiers using self-organizing maps
Clancy, T. C., & Khawar, A. (2009) · International Conference on Cognitive Radio Oriented Wireless Networks and Communications · doi:10.1109/crowncom.2009.5189050 · Cited by 17 · cited 1×
“Prior security threats to cognitive signal classifiers have been researched [25], [26], yet, the state of the art signal classification systems use deep learning techniques [8]–[13] whose vulnerabilities have not been studied extensively in”
BackgroundRecognizing FM, BPSK and 16-QAM using supervised and unsupervised learning techniques
Bari, M., Khawar, A., Doroslovacki, M. I., & Clancy, T. C. (2015) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/acssc.2015.7421104 · Cited by 13 · cited 1×
“Prior approaches to spectrum sensing were likelihood or feature based [2]–[7] while more recent approaches leverage the advances in deep neural networks (DNN) to operate directly on raw IQ samples [8]–[13].”
Background
Cited by
Adversarial Attacks in Modulation Recognition With Convolutional Neural Networks
Lin, Y., Zhao, H., Ma, X., Tu, Y., & Wang, M. (2021) · IEEE Transactions on Reliability · doi:10.1109/tr.2020.3032744 · Cited by 290
Machine learning in NextG networks via generative adversarial networks
Ayanoglu, E., Davaslioglu, K., & Sagduyu, Y. E. (2022) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2022.3153004 · Cited by 55
Metadata
- Authors
- Flowers, B., Buehrer, R. M., & Headley, W. C.
- Venue
- IEEE Transactions on Information Forensics and Security
- Published
- 2020
- Volume
- 15
- Pages
- 1102-1113
- ISSN
- 1556-6013 (ISSN-L)
- 1556-6021
- OpenAlex
- W2920763889
- Cited by
- 189
- References
- 32
- FWCI
- 10.02
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.