Browse by author
Lookup NU author(s): Dr Yachao Ran, Professor Gui Yun TianORCiD
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND).
© 2024 The Author(s). IET Wireless Sensor Systems published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology. Security problems loom big in the fast-growing world of Internet of Things (IoT) networks, which is characterised by unprecedented interconnectedness and data-driven innovation, due to the inherent susceptibility of wireless infrastructure. One of the most pressing concerns is user authentication, which was originally intended to prevent unwanted access to critical information but has since expanded to provide tailored service customisation. We suggest a Wi-Fi sensing-based physical layer authentication method for IoT networks to solve this problem. Our proposed method makes use of raw channel state information (CSI) data from Wi-Fi signals to create a hybrid deep-learning model that combines convolutional neural networks and long short-term memory networks. Rigorous testing yields an astonishing 99.97% accuracy rate, demonstrating the effectiveness of our CSI-based verification. This technology not only strengthens wireless network security but also prioritises efficiency and portability. The findings highlight the practicality of our proposed CSI-based physical layer authentication, which provides lightweight and precise protection for wireless networks in the IoT.
Author(s): Roopak M, Ran Y, Chen X, Tian GY, Parkinson S
Publication type: Article
Publication status: Published
Journal: IET Wireless Sensor Systems
Year: 2024
Volume: 14
Issue: 6
Pages: 441-450
Print publication date: 01/12/2024
Online publication date: 10/09/2024
Acceptance date: 24/08/2024
Date deposited: 24/09/2024
ISSN (print): 2043-6386
ISSN (electronic): 2043-6394
Publisher: John Wiley and Sons Inc.
URL: https://doi.org/10.1049/wss2.12093
DOI: 10.1049/wss2.12093
Data Access Statement: Data available on request due to privacy/ethical restrictions.
Altmetrics provided by Altmetric
