Poster: Breaking graph-based IoT malware detection systems using adversarial examples

Ahmed Abusnaina; Aminollah Khormali; Hisham Alasmary; Jeman Park; Afsah Anwar; Ulku Meteriz; Aziz Mohaisen

doi:10.1145/3317549.3326296

Poster: Breaking graph-based IoT malware detection systems using adversarial examples

Ahmed Abusnaina, Aminollah Khormali, Hisham Alasmary, Jeman Park, Afsah Anwar, Ulku Meteriz, Aziz Mohaisen

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

5 Citations (Scopus)

Abstract

The main goal of this study is to investigate the robustness of graphbased Deep Learning (DL) models used for Internet of Things (IoT) malware classification against Adversarial Learning (AL). We designed two approaches to craft adversarial IoT software, including Off-the-Shelf Adversarial Attack (OSAA) methods, using six different AL attack approaches, and Graph Embedding and Augmentation (GEA). The GEA approach aims to preserve the functionality and practicality of the generated adversarial sample through a careful embedding of a benign sample to a malicious one. Our evaluations demonstrate that OSAAs are able to achieve a misclassification rate (MR) of 100%. Moreover, we observed that the GEA approach is able to misclassify all IoT malware samples as benign.

Original language	English
Title of host publication	WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks
Publisher	Association for Computing Machinery, Inc
Pages	290-291
Number of pages	2
ISBN (Electronic)	9781450367264
DOIs	https://doi.org/10.1145/3317549.3326296
Publication status	Published - 15 May 2019
Event	12th Conference on Security and Privacy in Wireless and Mobile Networks, WiSec 2019 - Miami, United States Duration: 15 May 2019 → 17 May 2019

Publication series

Name	WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks

Conference

Conference	12th Conference on Security and Privacy in Wireless and Mobile Networks, WiSec 2019
Country/Territory	United States
City	Miami
Period	15/05/19 → 17/05/19

Bibliographical note

Publisher Copyright:
© 2019 Copyright held by the owner/author(s).

Keywords

Adversarial Learning
Deep Learning
Graph Analysis
Internet of Things
Malware Detection

Access to Document

10.1145/3317549.3326296

Cite this

Abusnaina, A., Khormali, A., Alasmary, H., Park, J., Anwar, A., Meteriz, U., & Mohaisen, A. (2019). Poster: Breaking graph-based IoT malware detection systems using adversarial examples. In WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks (pp. 290-291). (WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks). Association for Computing Machinery, Inc. https://doi.org/10.1145/3317549.3326296

Abusnaina, Ahmed ; Khormali, Aminollah ; Alasmary, Hisham et al. / Poster : Breaking graph-based IoT malware detection systems using adversarial examples. WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks. Association for Computing Machinery, Inc, 2019. pp. 290-291 (WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks).

@inproceedings{eda4d095a6e74b699d5f00da2eb58a66,

title = "Poster: Breaking graph-based IoT malware detection systems using adversarial examples",

abstract = "The main goal of this study is to investigate the robustness of graphbased Deep Learning (DL) models used for Internet of Things (IoT) malware classification against Adversarial Learning (AL). We designed two approaches to craft adversarial IoT software, including Off-the-Shelf Adversarial Attack (OSAA) methods, using six different AL attack approaches, and Graph Embedding and Augmentation (GEA). The GEA approach aims to preserve the functionality and practicality of the generated adversarial sample through a careful embedding of a benign sample to a malicious one. Our evaluations demonstrate that OSAAs are able to achieve a misclassification rate (MR) of 100%. Moreover, we observed that the GEA approach is able to misclassify all IoT malware samples as benign.",

keywords = "Adversarial Learning, Deep Learning, Graph Analysis, Internet of Things, Malware Detection",

author = "Ahmed Abusnaina and Aminollah Khormali and Hisham Alasmary and Jeman Park and Afsah Anwar and Ulku Meteriz and Aziz Mohaisen",

note = "Publisher Copyright: {\textcopyright} 2019 Copyright held by the owner/author(s).; 12th Conference on Security and Privacy in Wireless and Mobile Networks, WiSec 2019 ; Conference date: 15-05-2019 Through 17-05-2019",

year = "2019",

month = may,

day = "15",

doi = "10.1145/3317549.3326296",

language = "English",

series = "WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks",

publisher = "Association for Computing Machinery, Inc",

pages = "290--291",

booktitle = "WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks",

}

Abusnaina, A, Khormali, A, Alasmary, H, Park, J, Anwar, A, Meteriz, U & Mohaisen, A 2019, Poster: Breaking graph-based IoT malware detection systems using adversarial examples. in WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks. WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks, Association for Computing Machinery, Inc, pp. 290-291, 12th Conference on Security and Privacy in Wireless and Mobile Networks, WiSec 2019, Miami, United States, 15/05/19. https://doi.org/10.1145/3317549.3326296

Poster: Breaking graph-based IoT malware detection systems using adversarial examples. / Abusnaina, Ahmed; Khormali, Aminollah; Alasmary, Hisham et al.
WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks. Association for Computing Machinery, Inc, 2019. p. 290-291 (WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks).

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

TY - GEN

T1 - Poster

T2 - 12th Conference on Security and Privacy in Wireless and Mobile Networks, WiSec 2019

AU - Abusnaina, Ahmed

AU - Khormali, Aminollah

AU - Alasmary, Hisham

AU - Park, Jeman

AU - Anwar, Afsah

AU - Meteriz, Ulku

AU - Mohaisen, Aziz

PY - 2019/5/15

Y1 - 2019/5/15

N2 - The main goal of this study is to investigate the robustness of graphbased Deep Learning (DL) models used for Internet of Things (IoT) malware classification against Adversarial Learning (AL). We designed two approaches to craft adversarial IoT software, including Off-the-Shelf Adversarial Attack (OSAA) methods, using six different AL attack approaches, and Graph Embedding and Augmentation (GEA). The GEA approach aims to preserve the functionality and practicality of the generated adversarial sample through a careful embedding of a benign sample to a malicious one. Our evaluations demonstrate that OSAAs are able to achieve a misclassification rate (MR) of 100%. Moreover, we observed that the GEA approach is able to misclassify all IoT malware samples as benign.

AB - The main goal of this study is to investigate the robustness of graphbased Deep Learning (DL) models used for Internet of Things (IoT) malware classification against Adversarial Learning (AL). We designed two approaches to craft adversarial IoT software, including Off-the-Shelf Adversarial Attack (OSAA) methods, using six different AL attack approaches, and Graph Embedding and Augmentation (GEA). The GEA approach aims to preserve the functionality and practicality of the generated adversarial sample through a careful embedding of a benign sample to a malicious one. Our evaluations demonstrate that OSAAs are able to achieve a misclassification rate (MR) of 100%. Moreover, we observed that the GEA approach is able to misclassify all IoT malware samples as benign.

KW - Adversarial Learning

KW - Deep Learning

KW - Graph Analysis

KW - Internet of Things

KW - Malware Detection

UR - http://www.scopus.com/inward/record.url?scp=85066749109&partnerID=8YFLogxK

U2 - 10.1145/3317549.3326296

DO - 10.1145/3317549.3326296

M3 - Conference contribution

AN - SCOPUS:85066749109

T3 - WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks

SP - 290

EP - 291

BT - WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks

PB - Association for Computing Machinery, Inc

Y2 - 15 May 2019 through 17 May 2019

ER -

Abusnaina A, Khormali A, Alasmary H, Park J, Anwar A, Meteriz U et al. Poster: Breaking graph-based IoT malware detection systems using adversarial examples. In WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks. Association for Computing Machinery, Inc. 2019. p. 290-291. (WiSec 2019 - Proceedings of the 2019 Conference on Security and Privacy in Wireless and Mobile Networks). doi: 10.1145/3317549.3326296

Poster: Breaking graph-based IoT malware detection systems using adversarial examples

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this