DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks

Rui Ning; Cong Wang; Chunsheng Xin; Jiang Li; Hongyi Wu

doi:10.1109/PERCOM.2018.8444573

DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks

Rui Ning
, Cong Wang
, Chunsheng Xin
, Jiang Li
, Hongyi Wu

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

29 Scopus citations

Abstract

In this paper, we report a newfound vulnerability on smartphones due to the malicious use of unsupervised sensor data. We demonstrate that an attacker can train deep Convolutional Neural Networks (CNN) by using magnetometer or orientation data to effectively infer the Apps and their usage information on a smartphone with an accuracy of over 80%. Furthermore, we show that such attacks can become even worse if sophisticated attackers exploit motion sensors to cluster the magnetometer or orientation data, improving the accuracy to as high as 98%. To mitigate such attacks, we propose a noise injection scheme that can effectively reduce the App sniffing accuracy to only 15% and at the same time has negligible effect on benign Apps.

Original language	English (US)
Title of host publication	Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Print)	9781538632246
DOIs	https://doi.org/10.1109/PERCOM.2018.8444573
State	Published - Aug 22 2018
Externally published	Yes
Event	16th IEEE International Conference on Pervasive Computing and Communications, PerCom 2018 - Athens, Greece Duration: Mar 19 2018 → Mar 23 2018

Publication series

Name	2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018

Conference

Conference	16th IEEE International Conference on Pervasive Computing and Communications, PerCom 2018
Country/Territory	Greece
City	Athens
Period	3/19/18 → 3/23/18

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Computer Science Applications

Access to Document

10.1109/PERCOM.2018.8444573

Cite this

Ning, R., Wang, C., Xin, C., Li, J., & Wu, H. (2018). DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks. In Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018 Article 8444573 (2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PERCOM.2018.8444573

DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks. / Ning, Rui; Wang, Cong; Xin, Chunsheng et al.
Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018. Institute of Electrical and Electronics Engineers Inc., 2018. 8444573 (2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018).

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

Ning, R, Wang, C, Xin, C, Li, J & Wu, H 2018, DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks. in Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018., 8444573, 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018, Institute of Electrical and Electronics Engineers Inc., 16th IEEE International Conference on Pervasive Computing and Communications, PerCom 2018, Athens, Greece, 3/19/18. https://doi.org/10.1109/PERCOM.2018.8444573

Ning R, Wang C, Xin C, Li J, Wu H. DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks. In Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018. Institute of Electrical and Electronics Engineers Inc. 2018. 8444573. (2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018). doi: 10.1109/PERCOM.2018.8444573

Ning, Rui ; Wang, Cong ; Xin, Chunsheng et al. / DeepMag : Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks. Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018. Institute of Electrical and Electronics Engineers Inc., 2018. (2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018).

@inproceedings{9863ebdcd1aa4dd783fb63bd99f3367f,

title = "DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks",

abstract = "In this paper, we report a newfound vulnerability on smartphones due to the malicious use of unsupervised sensor data. We demonstrate that an attacker can train deep Convolutional Neural Networks (CNN) by using magnetometer or orientation data to effectively infer the Apps and their usage information on a smartphone with an accuracy of over 80\%. Furthermore, we show that such attacks can become even worse if sophisticated attackers exploit motion sensors to cluster the magnetometer or orientation data, improving the accuracy to as high as 98\%. To mitigate such attacks, we propose a noise injection scheme that can effectively reduce the App sniffing accuracy to only 15\% and at the same time has negligible effect on benign Apps.",

author = "Rui Ning and Cong Wang and Chunsheng Xin and Jiang Li and Hongyi Wu",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 16th IEEE International Conference on Pervasive Computing and Communications, PerCom 2018 ; Conference date: 19-03-2018 Through 23-03-2018",

year = "2018",

month = aug,

day = "22",

doi = "10.1109/PERCOM.2018.8444573",

language = "English (US)",

isbn = "9781538632246",

series = "2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018",

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

booktitle = "Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018",

}

TY - GEN

T1 - DeepMag

T2 - 16th IEEE International Conference on Pervasive Computing and Communications, PerCom 2018

AU - Ning, Rui

AU - Wang, Cong

AU - Xin, Chunsheng

AU - Li, Jiang

AU - Wu, Hongyi

PY - 2018/8/22

Y1 - 2018/8/22

N2 - In this paper, we report a newfound vulnerability on smartphones due to the malicious use of unsupervised sensor data. We demonstrate that an attacker can train deep Convolutional Neural Networks (CNN) by using magnetometer or orientation data to effectively infer the Apps and their usage information on a smartphone with an accuracy of over 80%. Furthermore, we show that such attacks can become even worse if sophisticated attackers exploit motion sensors to cluster the magnetometer or orientation data, improving the accuracy to as high as 98%. To mitigate such attacks, we propose a noise injection scheme that can effectively reduce the App sniffing accuracy to only 15% and at the same time has negligible effect on benign Apps.

AB - In this paper, we report a newfound vulnerability on smartphones due to the malicious use of unsupervised sensor data. We demonstrate that an attacker can train deep Convolutional Neural Networks (CNN) by using magnetometer or orientation data to effectively infer the Apps and their usage information on a smartphone with an accuracy of over 80%. Furthermore, we show that such attacks can become even worse if sophisticated attackers exploit motion sensors to cluster the magnetometer or orientation data, improving the accuracy to as high as 98%. To mitigate such attacks, we propose a noise injection scheme that can effectively reduce the App sniffing accuracy to only 15% and at the same time has negligible effect on benign Apps.

UR - https://www.scopus.com/pages/publications/85053446425

UR - https://www.scopus.com/pages/publications/85053446425#tab=citedBy

U2 - 10.1109/PERCOM.2018.8444573

DO - 10.1109/PERCOM.2018.8444573

M3 - Conference contribution

AN - SCOPUS:85053446425

SN - 9781538632246

T3 - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018

BT - Proceedings - 2018 IEEE International Conference on Pervasive Computing and Communications, PerCom 2018

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 19 March 2018 through 23 March 2018

ER -

DeepMag: Sniffing Mobile Apps in Magnetic Field through Deep Convolutional Neural Networks

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this