A Defensive Strategy Against Beam Training Attack in 5G mmWave Networks for Manufacturing

Dinh-Van, Son and Hoang, Tiep M. and Cebecioglu, Berna Bulut and Fowler, Daniel S. and Mo, Yuen Kwan and Higgins, Matthew (2023) A Defensive Strategy Against Beam Training Attack in 5G mmWave Networks for Manufacturing. IEEE Transactions on Information Forensics and Security, 18. pp. 2204-2217. ISSN 1556-6013

Preview

Text Defensive-strategy-against-beam-training-attack-networks-manufacturing-BCU.pdf - Accepted Version Download (27MB)

Abstract

Millimeter-wave (mmWave) carriers are an essential building block of fifth-generation (5G) systems. Satisfactory performance of the communications over the mmWave spectrum requires an alignment between the signal beam of the transmitter and receiver, achieved via beam training protocols. Nevertheless, beam training is vulnerable to jamming attacks, where the attacker intends to send jamming signals over different spatial directions to confuse legitimate nodes. This paper focuses on defending against this attack in smart factories where a moving Automated Guided Vehicle (AGV) communicates with a base station via a mmWave carrier. We introduce a defensive strategy to cope with jamming attacks, including two stages: jamming detection and jamming mitigation. Developed based on autoencoders, both algorithms can learn the characteristics/features of the received signals at the AGV. They can be employed consecutively before performing the downlink data transmission. In particular, once a jamming attack is identified, the jamming mitigation can be utilized to retrieve the corrupted received signal strength vector, allowing a better decision during the beam training operation. In addition, the proposed algorithm is straightforward and fully compliant with the existing beam training protocols in 5G New Radio. The numerical results show that not only the proposed defensive strategy can capture more than 80% of attack events, but it also improves the average signal-to-interference-plus-noise-ratio significantly, i.e., up to 15 dB.

Item Type:	Article
Identification Number:	10.1109/TIFS.2023.3265341
Dates:	Date Event 1 April 2023 Accepted 6 April 2023 Published Online
Uncontrolled Keywords:	Attack detection, beam training, beam training attack, machine learning, mmWave, PHY-layer security, 5G
Subjects:	CAH11 - computing > CAH11-01 - computing > CAH11-01-01 - computer science
Divisions:	Faculty of Computing, Engineering and the Built Environment > College of Computing
Depositing User:	Gemma Tonks
Date Deposited:	02 Jul 2024 13:24
Last Modified:	02 Jul 2024 13:24
URI:	https://www.open-access.bcu.ac.uk/id/eprint/15626

Actions (login required)

View Item