Secure Energy Efficiency in Communication Systems Based on Deep Learning

Zou Xiangyu,Huang Chongwen,Xu Yongjun,Yang Zhaohui,Cao Yue
DOI: https://doi.org/10.11999/JEIT211611
2022-01-01
Abstract:The secure transmission for Reconfigurable Intelligent Surface (RIS) assisted wireless communication systems is investigated in this paper. In the studied model, one user connects to the access point via a RIS while an eavesdropper eavesdrops on the signal sent from the user to the access point. Therefore, it is necessary to design an appropriate RIS reflection vector to solve the eavesdropping problem. This problem is formulated as an optimization problem whose goal is to maximize the secure energy efficiency which is defined via jointlyoptimizing the RIS reflection reflector as well as the number of RIS elements, which results in a non-convex optimization problem that is intractable to solve by the traditional methods. To tackle this issue, a new algorithm by leveraging the advance of the emerging Deep Learning (DL) technique is proposed so as to find the near optimal RIS reflection vector and determine the near optimal number of RIS reflection elements. Simulation results show that the proposed method reaches 96% of the near optimal secure energy efficiency of the genie-aided algorithm.
What problem does this paper attempt to address?