Collaborative Inference Over Wireless Channels With Feature Differential Privacy
IEEE Journal on Selected Areas in Communications · 2025-12 · 43 (12) · 4027-4041
DOI: 10.1109/jsac.2025.3642520 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Journal on Selected Areas in Communications, 2025-12. Verfasst von Mohamed Seif und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 30 Literaturangaben. Gefördert von: Air Force Office of Scientific Research, National Science Foundation, Innovation Grant from Princeton NextG.
Abstract (Originalsprache)
Collaborative inference among multiple wireless edge devices has the potential to significantly enhance Artificial Intelligence (AI) applications, particularly for sensing and computer vision. This approach typically involves a three-stage process: 1) data acquisition through sensing, 2) feature extraction, and 3) feature encoding for transmission. However, transmitting the extracted features poses a significant privacy risk, as sensitive personal data can be exposed during the process. To address this challenge, we propose a novel privacy-preserving collaborative inference mechanism, wherein each edge device in the network secures the privacy of extracted features before transmitting them to a central server for inference. Our approach is designed to achieve two primary objectives: 1) reducing communication overhead and 2) ensuring strict privacy guarantees during feature transmission, while maintaining effective inference performance. Additionally, we introduce an over-the-air pooling scheme specifically designed for classification tasks, which provides formal guarantees on the privacy of transmitted features and establishes a lower bound on classification accuracy.
Förderung
- Air Force Office of Scientific Research
- National Science Foundation
- Innovation Grant from Princeton NextG
Zitieren
APA
Seif, M., Nie, Y., Goldsmith, A. J., & Poor, H. V. (2025). Collaborative Inference Over Wireless Channels With Feature Differential Privacy. IEEE Journal on Selected Areas in Communications, 43(12), 4027-4041. https://doi.org/10.1109/jsac.2025.3642520
Vancouver
Seif M, Nie Y, Goldsmith AJ, Poor HV. Collaborative Inference Over Wireless Channels With Feature Differential Privacy. IEEE Journal on Selected Areas in Communications. 2025;43(12):4027-4041. doi:10.1109/jsac.2025.3642520
BibTeX
@article{Seif2025,
author = {Seif, Mohamed and Nie, Yuqi and Goldsmith, Andrea J. and Poor, H. Vincent},
title = {Collaborative Inference Over Wireless Channels With Feature Differential Privacy},
journal = {IEEE Journal on Selected Areas in Communications},
year = {2025},
volume = {43},
number = {12},
pages = {4027--4041},
doi = {10.1109/jsac.2025.3642520}
}
EUVIMED