/MagBackdoor

MagBackdoor: Beware of Your Loudspeaker as A Backdoor For Magnetic Injection Attacks

MagBackdoor

MagBackdoor: Beware of Your Loudspeaker as A Backdoor For Magnetic Injection Attacks in IEEE Symposium on Security and Privacy 2023

Citing

If you are inspired by this paper, please cite

@INPROCEEDINGS {Tiantian_MagBackdoor,
author = {Tiantian Liu and Feng Lin and Zhangsen Wang and Chao Wang and Zhongjie Ba and Li Lu and Wenyao Xu and Kui Ren},
booktitle = {2023 IEEE Symposium on Security and Privacy (SP)},
title = {MagBackdoor: Beware of Your Loudspeaker as a Backdoor for Magnetic Injection Attacks},
year = {2023},
volume = {},
issn = {},
pages = {3416-3431},
abstract = {An audio system containing loudspeakers and microphones is the fundamental hardware for voice-enabled devices, enabling voice interaction with mobile applications and smart homes. This paper presents MagBackdoor, the first magnetic field attack that injects malicious commands via a loudspeaker-based backdoor of the audio system, compromising the linked voice interaction system. MagBackdoor focuses on the magnetic threat on loudspeakers and manipulates their sound production stealthily. Consequently, the microphone will inevitably pick up malicious sound generated by the attacked speaker, due to the closely packed arrangement of internal audio systems. To prove the feasibility of MagBackdoor, we conduct comprehensive simulations and experiments. This study further models the mechanism by which an external magnetic field excites the sound production of loudspeakers, giving theoretical guidance to MagBackdoor. Aiming at stealthy magnetic attacks in real-world scenarios, we self-design a prototype that can emit magnetic fields modulated by voice commands. We implement MagBackdoor and evaluate it across a wide range of smart devices involving 16 smartphones, four laptops, two tablets, and three smart speakers, achieving an average 95% injection success rate with high-quality injected acoustic signals.},
keywords = {},
doi = {10.1109/SP46215.2023.00132},
url = {https://doi.ieeecomputersociety.org/10.1109/SP46215.2023.00132},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
month = {may}
}