AF Relaying Secrecy Performance Prediction for 6G Mobile Communication Networks in Industry 5.0

Lingwei Xu, Xinpeng Zhou, Ye Tao, Xu Yu, Miao Yu, Fazlullah Khan

Research output: Journal PublicationArticlepeer-review

28 Citations (Scopus)

Abstract

Industry 5.0 has developed in full swing, and accelerated the process of the sixth-generation (6G) mobile communication. Physical layer security is important for complex 6G mobile communication networks. To process active complex events in 6G mobile cooperative networks, predicting secrecy performance in time is essential for the mobile communication quality evaluation. Using amplify-and-forward (AF) relaying, we propose a transmit antenna selection (TAS) based secrecy scheme in this article. To analyze the security of 6G mobile cooperative networks, signal-to-noise ratio of the end-to-end link is used to derive the novel expressions for secrecy outage probability (SOP). The theoretical results are confirmed by simulation results. Then, we use SOP as the important merit to evaluate the secrecy performance, and set up the dataset. To achieve secrecy performance prediction, a convolutional neural network (CNN) based SOP prediction algorithm is proposed. The designed CNN model has five convolution layers, which all use the same convolution in the padding and do not change data size. For this improved CNN structure, we adopt the idea of SqueezeNet, which belongs to the lightweight CNN. The improved CNN model can greatly reduce the parameters and network complexity on the premise of ensuring the prediction accuracy. We also examine the following state-of-the-art techniques, first, Elman, second, InceptionNet, third, deep neural network (DNN), and fourth, support vector machine methods. The proposed CNN algorithm can achieve better SOP prediction results than other existing methods. In particular, compared with DNN method, the prediction accuracy is increased by 66.7%.

Original languageEnglish
Pages (from-to)5485-5493
Number of pages9
JournalIEEE Transactions on Industrial Informatics
Volume18
Issue number8
DOIs
Publication statusPublished - 1 Aug 2022
Externally publishedYes

Keywords

  • Improved convolutional neural network (CNN)
  • physical layer security
  • secrecy performance prediction
  • sixth-generation (6G) mobile communication

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Information Systems
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'AF Relaying Secrecy Performance Prediction for 6G Mobile Communication Networks in Industry 5.0'. Together they form a unique fingerprint.

Cite this