Securing Contrastive mmWave-based Human Activity Recognition against Adversarial Label Flipping. Singha, A., Bi, Z., Li, T., Chen, Y., & Zhang, Y. In Proceedings of the 17th ACM Conference on Security and Privacy in Wireless and Mobile Networks, of WiSec '24, pages 31–41, New York, NY, USA, May, 2024. Association for Computing Machinery. ![Securing Contrastive mmWave-based Human Activity Recognition against Adversarial Label Flipping [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex 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.
@inproceedings{singha_securing_2024,
address = {New York, NY, USA},
series = {{WiSec} '24},
title = {Securing {Contrastive} {mmWave}-based {Human} {Activity} {Recognition} against {Adversarial} {Label} {Flipping}},
isbn = {9798400705823},
url = {https://dl.acm.org/doi/10.1145/3643833.3656123},
doi = {10.1145/3643833.3656123},
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.},
urldate = {2024-10-06},
booktitle = {Proceedings of the 17th {ACM} {Conference} on {Security} and {Privacy} in {Wireless} and {Mobile} {Networks}},
publisher = {Association for Computing Machinery},
author = {Singha, Amit and Bi, Ziqian and Li, Tao and Chen, Yimin and Zhang, Yanchao},
month = may,
year = {2024},
pages = {31--41},
}
Downloads: 0
{"_id":"MY7CC7GPPYib2K3Wo","bibbaseid":"singha-bi-li-chen-zhang-securingcontrastivemmwavebasedhumanactivityrecognitionagainstadversariallabelflipping-2024","author_short":["Singha, A.","Bi, Z.","Li, T.","Chen, Y.","Zhang, Y."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"New York, NY, USA","series":"WiSec '24","title":"Securing Contrastive mmWave-based Human Activity Recognition against Adversarial Label Flipping","isbn":"9798400705823","url":"https://dl.acm.org/doi/10.1145/3643833.3656123","doi":"10.1145/3643833.3656123","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.","urldate":"2024-10-06","booktitle":"Proceedings of the 17th ACM Conference on Security and Privacy in Wireless and Mobile Networks","publisher":"Association for Computing Machinery","author":[{"propositions":[],"lastnames":["Singha"],"firstnames":["Amit"],"suffixes":[]},{"propositions":[],"lastnames":["Bi"],"firstnames":["Ziqian"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Tao"],"suffixes":[]},{"propositions":[],"lastnames":["Chen"],"firstnames":["Yimin"],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["Yanchao"],"suffixes":[]}],"month":"May","year":"2024","pages":"31–41","bibtex":"@inproceedings{singha_securing_2024,\n\taddress = {New York, NY, USA},\n\tseries = {{WiSec} '24},\n\ttitle = {Securing {Contrastive} {mmWave}-based {Human} {Activity} {Recognition} against {Adversarial} {Label} {Flipping}},\n\tisbn = {9798400705823},\n\turl = {https://dl.acm.org/doi/10.1145/3643833.3656123},\n\tdoi = {10.1145/3643833.3656123},\n\tabstract = {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.},\n\turldate = {2024-10-06},\n\tbooktitle = {Proceedings of the 17th {ACM} {Conference} on {Security} and {Privacy} in {Wireless} and {Mobile} {Networks}},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Singha, Amit and Bi, Ziqian and Li, Tao and Chen, Yimin and Zhang, Yanchao},\n\tmonth = may,\n\tyear = {2024},\n\tpages = {31--41},\n}\n\n","author_short":["Singha, A.","Bi, Z.","Li, T.","Chen, Y.","Zhang, Y."],"key":"singha_securing_2024","id":"singha_securing_2024","bibbaseid":"singha-bi-li-chen-zhang-securingcontrastivemmwavebasedhumanactivityrecognitionagainstadversariallabelflipping-2024","role":"author","urls":{"Paper":"https://dl.acm.org/doi/10.1145/3643833.3656123"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/8994433/collections/66SDI5S7/items?key=ozCmKMmrE26WwEN9PAdOLaiV&format=bibtex&limit=100","dataSources":["yM68t35RzDgJZajbn","nXz2cN22CbQDukEnJ","Rxy8umoNP78qdq56v"],"keywords":[],"search_terms":["securing","contrastive","mmwave","based","human","activity","recognition","against","adversarial","label","flipping","singha","bi","li","chen","zhang"],"title":"Securing Contrastive mmWave-based Human Activity Recognition against Adversarial Label Flipping","year":2024}