Backdoor Attacks Against Deep Learning-based Massive MIMO Localization

被引：2

作者：

Zhao, Tianya ^{[1
]}

Wang, Xuyu ^{[1
]}

Mao, Shiwen ^{[2
]}

机构：

[1] Florida Int Univ, Knight Fdn, Sch Comp & Informat Sci, Miami, FL 33199 USA

[2] Auburn Univ, Dept Elect & Comp Engn, Auburn, AL 36849 USA

来源：

IEEE CONFERENCE ON GLOBAL COMMUNICATIONS, GLOBECOM | 2023年

关键词：

Backdoor Attack; Deep Learning; Massive MIMO; Wireless Localization;

D O I：

10.1109/GLOBECOM54140.2023.10437534

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Millimeter wave (mmWave) communications and massive MIMO play crucial roles in the development of future wireless systems. In addition to offering high data rates, these technologies enable the realization of high-precision localization systems, especially in complicated indoor rich multi-path environments without GPS coverage. While deep neural networks (DNNs) enable high accuracy in fingerprint-based indoor localization, their implementations also introduce security problems. In the field of computer vision, backdoor attacks have proven to be able to effectively deceive models using specific or imperceptible triggers. In this paper, we study the impact of backdoor attacks on 5G massive MIMO localization systems in both indoor and outdoor environments. Two different triggers are investigated: the one-pixel trigger (visible) and the random noise trigger (invisible). We evaluate the localization systems using a public dataset and demonstrate that DNN-based localization systems are vulnerable to backdoor attacks.

引用

页码：2796 / 2801

页数：6

共 21 条

[1] Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey [J].

Akhtar, Naveed ;

Mian, Ajmal .

IEEE ACCESS, 2018, 6 :14410-14430

[2] Contribution of the Zubair source rocks to the generation and expulsion of oil to the reservoirs of the Mesopotamian Basin, Southern Iraq [J].

Al-Khafaji, Amer Jassim ;

Sadooni, Fadhil ;

Hindi, Mohammed Hadi .

PETROLEUM SCIENCE AND TECHNOLOGY, 2019, 37 (08) :940-949

[3] Robust Adversarial Attacks Against DNN-Based Wireless Communication Systems [J].

Bahramali, Alireza ;

Nasr, Milad ;

Houmansadr, Amir ;

Goeckel, Dennis ;

Towsley, Don .

CCS '21: PROCEEDINGS OF THE 2021 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, 2021, :126-140

[4]

Boora U, 2022, IEEE ICC, P4866, DOI 10.1109/ICC45855.2022.9838836

[5] Millimeter-Wave Massive MIMO Communication for Future Wireless Systems: A Survey [J].

Busari, Sherif Adeshina ;

Huq, Kazi Mohammed Saidul ;

Mumtaz, Shahid ;

Dai, Linglong ;

Rodriguez, Jonathan .

IEEE COMMUNICATIONS SURVEYS AND TUTORIALS, 2018, 20 (02) :836-869

[6] Elevated c-kit expression in failed autologous arteriovenous fistulas in end-stage renal disease patients [J].

Du, Jing ;

Li, Yang ;

Jia, Xiaoyan ;

Kong, Xianglei ;

Liang, Liming ;

Wang, Dongmei ;

Li, Anzhuang ;

Chen, Qinlan ;

Su, Hong ;

Li, Wenbin ;

Xu, Dongmei .

JOURNAL OF VASCULAR ACCESS, 2024, 25 (02) :423-431

[7]

Gao Yansong, 2020, arXiv

[8]

Hejazi F., 2021, IEEE INFOCOM 2021 IE, P1

[9] IS-WARS: Intelligent and Stealthy Adversarial Attack to Wi-Fi-Based Human Activity Recognition Systems [J].

Huang, Pei ;

Zhang, Xiaonan ;

Yu, Sihan ;

Guo, Linke .

IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING, 2022, 19 (06) :3899-3912

[10] Universal adversarial perturbations [J].

Moosavi-Dezfooli, Seyed-Mohsen ;

Fawzi, Alhussein ;

Fawzi, Omar ;

Frossard, Pascal .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :86-94

← 1 2 3 →