Black-box adversarial machine learning attack on network traffic classification

Muhammad Usama; Adnan Qayyum; Junaid Qadir; Ala Al-Fuqaha

doi:10.1109/IWCMC.2019.8766505

Black-box adversarial machine learning attack on network traffic classification

Muhammad Usama
, Adnan Qayyum
, Junaid Qadir
, Ala Al-Fuqaha

Information Technology University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

53 Citations (Scopus)

Abstract

Deep machine learning techniques have shown promising results in network traffic classification, however, the robustness of these techniques under adversarial threats is still in question. Deep machine learning models are found vulnerable to small carefully crafted adversarial perturbations posing a major question on the performance of deep machine learning techniques. In this paper, we propose a black-box adversarial attack on network traffic classification. The proposed attack successfully evades deep machine learning-based classifiers which highlights the potential security threat of using deep machine learning techniques to realize autonomous networks.

Original language	English
Title of host publication	2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	84-89
Number of pages	6
ISBN (Electronic)	9781538677476
DOIs	https://doi.org/10.1109/IWCMC.2019.8766505
Publication status	Published - Jun 2019
Event	15th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2019 - Tangier, Morocco Duration: 24 Jun 2019 → 28 Jun 2019

Publication series

Name	2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019

Conference

Conference	15th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2019
Country/Territory	Morocco
City	Tangier
Period	24/06/19 → 28/06/19

Keywords

Adversarial machine learning
Black-box adversarial attack
Network traffic classification

Access to Document

10.1109/IWCMC.2019.8766505

Cite this

Usama, M., Qayyum, A., Qadir, J., & Al-Fuqaha, A. (2019). Black-box adversarial machine learning attack on network traffic classification. In 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019 (pp. 84-89). Article 8766505 (2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IWCMC.2019.8766505

Usama, Muhammad ; Qayyum, Adnan ; Qadir, Junaid et al. / Black-box adversarial machine learning attack on network traffic classification. 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 84-89 (2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019).

@inproceedings{1439390430d744fca48422e27e0633da,

title = "Black-box adversarial machine learning attack on network traffic classification",

abstract = "Deep machine learning techniques have shown promising results in network traffic classification, however, the robustness of these techniques under adversarial threats is still in question. Deep machine learning models are found vulnerable to small carefully crafted adversarial perturbations posing a major question on the performance of deep machine learning techniques. In this paper, we propose a black-box adversarial attack on network traffic classification. The proposed attack successfully evades deep machine learning-based classifiers which highlights the potential security threat of using deep machine learning techniques to realize autonomous networks.",

keywords = "Adversarial machine learning, Black-box adversarial attack, Network traffic classification",

author = "Muhammad Usama and Adnan Qayyum and Junaid Qadir and Ala Al-Fuqaha",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 15th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2019 ; Conference date: 24-06-2019 Through 28-06-2019",

year = "2019",

month = jun,

doi = "10.1109/IWCMC.2019.8766505",

language = "English",

series = "2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "84--89",

booktitle = "2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019",

address = "United States",

}

Usama, M, Qayyum, A, Qadir, J & Al-Fuqaha, A 2019, Black-box adversarial machine learning attack on network traffic classification. in 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019., 8766505, 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019, Institute of Electrical and Electronics Engineers Inc., pp. 84-89, 15th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2019, Tangier, Morocco, 24/06/19. https://doi.org/10.1109/IWCMC.2019.8766505

Black-box adversarial machine learning attack on network traffic classification. / Usama, Muhammad; Qayyum, Adnan; Qadir, Junaid et al.
2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 84-89 8766505 (2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Black-box adversarial machine learning attack on network traffic classification

AU - Usama, Muhammad

AU - Qayyum, Adnan

AU - Qadir, Junaid

AU - Al-Fuqaha, Ala

PY - 2019/6

Y1 - 2019/6

N2 - Deep machine learning techniques have shown promising results in network traffic classification, however, the robustness of these techniques under adversarial threats is still in question. Deep machine learning models are found vulnerable to small carefully crafted adversarial perturbations posing a major question on the performance of deep machine learning techniques. In this paper, we propose a black-box adversarial attack on network traffic classification. The proposed attack successfully evades deep machine learning-based classifiers which highlights the potential security threat of using deep machine learning techniques to realize autonomous networks.

AB - Deep machine learning techniques have shown promising results in network traffic classification, however, the robustness of these techniques under adversarial threats is still in question. Deep machine learning models are found vulnerable to small carefully crafted adversarial perturbations posing a major question on the performance of deep machine learning techniques. In this paper, we propose a black-box adversarial attack on network traffic classification. The proposed attack successfully evades deep machine learning-based classifiers which highlights the potential security threat of using deep machine learning techniques to realize autonomous networks.

KW - Adversarial machine learning

KW - Black-box adversarial attack

KW - Network traffic classification

UR - https://www.scopus.com/pages/publications/85073905957

U2 - 10.1109/IWCMC.2019.8766505

DO - 10.1109/IWCMC.2019.8766505

M3 - Conference contribution

AN - SCOPUS:85073905957

T3 - 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019

SP - 84

EP - 89

BT - 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 15th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2019

Y2 - 24 June 2019 through 28 June 2019

ER -

Usama M, Qayyum A, Qadir J, Al-Fuqaha A. Black-box adversarial machine learning attack on network traffic classification. In 2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 84-89. 8766505. (2019 15th International Wireless Communications and Mobile Computing Conference, IWCMC 2019). doi: 10.1109/IWCMC.2019.8766505

Black-box adversarial machine learning attack on network traffic classification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this