Securing Contrastive mmWave-based Human Activity Recognition against Adversarial Label Flipping

arXiv:2608.04029 · cs.CR · Submitted 2026-08-01 · Read on arXiv

Amit Singha, Ziqian Bi, Tao Li, Yimin Chen, Yanchao Zhang

cs.CR

Submitted: 2026-08-01

Comments: 11 pages, 18 figures. Published in Proceedings of the 17th ACM Conference on Security and Privacy in Wireless and Mobile Networks (WiSec '24)

Journal ref: Proceedings of the 17th ACM Conference on Security and Privacy in Wireless and Mobile Networks (WiSec '24), Seoul, Republic of Korea, 2024, pp. 31-41

DOI: 10.1145/3643833.3656123

License: http://creativecommons.org/licenses/by/4.0/

The gist: Wireless Human Activity Recognition (HAR), leveraging their non-intrusive nature, has the potential to revolutionize various sectors, including healthcare, virtual reality, and surveillance.

Terminology

Abstract

Wireless Human Activity Recognition (HAR), leveraging their non-intrusive nature, has the potential to revolutionize various sectors, including healthcare, virtual reality, and surveillance. The advent of millimeter wave (mmWave) technology has significantly enhanced the capabilities of wireless HAR systems. This paper presents the first systematic study on the vulnerabilities of mmWave-based HAR to label flipping poisoning attacks in the context of supervised contrastive learning. We identify three label poisoning attacks on the contrastive mmWave-based HAR and propose corresponding countermeasures. The efficacy of the attacks and also our countermeasures are experimentally validated on a prototype system. The attacks and countermeasures can be easily extended to other wireless HAR systems, thereby promoting security considerations in system design and deployment.

Sources

Related papers