Adversarial Attack Detection and Mitigation in Next-Generation Intelligent Consumer Electronics
IEEE Transactions on Consumer Electronics · 2026-05 · 72 (2) · 4671-4682
DOI: 10.1109/tce.2026.3684061 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Consumer Electronics, 2026-05. Verfasst von Basudeb Bera und 3 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Gefördert von: A*STAR, CISCO Systems (USA) Pte. Ltd, National University of Singapore through the Cisco-NUS Accelerated Digital Economy Corporate Laboratory.
Abstract (Originalsprache)
In next-generation consumer electronics (NGCE), advanced CE devices heavily rely on machine learning and artificial intelligence (AI), where data and models are the basic building blocks for making automated decisions and optimized services. Recently, generative adversarial networks has led to model inversion as well as model/data poisoning attacks in NGCE that can significantly degrade the model performance and accuracy. In addition, the majority of the CE data are collected through vulnerable wireless public channels, posing significant security challenges to data privacy, integrity, and confidentiality. However, traditional authentication schemes relying on the discrete logarithm problem (DLP) have been utilized for securing CE’s data. With the advancement of quantum computing and the adaptation of the quantum Shor’s algorithm, these schemes become no more secure and therefore vulnerable to quantum attacks. Moreover, the wireless CE applications are vulnerable to impersonation attacks by malicious entities. To overcome these challenges, we propose a radio frequency fingerprint (RFF)-based adversarial attack detection and mitigation with a lightweight quantum-secure authentication and key agreement protocol for securing CE communication. We utilize a vector similarity search technique for attack detection and unauthorized access using the CE’s RFF. A comprehensive performance analysis and comparative analysis presents its efficiency and scalability for CE applications. A testbed experiment using Raspberry Pi, verification using the Scyther tool, and performance under unknown attacks shows the practicability, correctness, and robustness of the proposed scheme. The network performance using NS-3 highlights the network performance of the proposed scheme compared to the existing schemes.
Förderung
- A*STAR, CISCO Systems (USA) Pte. Ltd
- National University of Singapore through the Cisco-NUS Accelerated Digital Economy Corporate Laboratory
Zitieren
APA
Bera, B., Bera, S., Liyanage, M., & Sikdar, B. (2026). Adversarial Attack Detection and Mitigation in Next-Generation Intelligent Consumer Electronics. IEEE Transactions on Consumer Electronics, 72(2), 4671-4682. https://doi.org/10.1109/tce.2026.3684061
Vancouver
Bera B, Bera S, Liyanage M, Sikdar B. Adversarial Attack Detection and Mitigation in Next-Generation Intelligent Consumer Electronics. IEEE Transactions on Consumer Electronics. 2026;72(2):4671-4682. doi:10.1109/tce.2026.3684061
BibTeX
@article{Bera2026,
author = {Bera, Basudeb and Bera, Samaresh and Liyanage, Madhusanka and Sikdar, Biplab},
title = {Adversarial Attack Detection and Mitigation in Next-Generation Intelligent Consumer Electronics},
journal = {IEEE Transactions on Consumer Electronics},
year = {2026},
volume = {72},
number = {2},
pages = {4671--4682},
doi = {10.1109/tce.2026.3684061}
}
EUVIMED