Secure Beamforming Optimization for IRS-NOMA Networks via Artificial Jamming

Abstract

In this paper, we incorporate intelligent reflecting surface (IRS) into non-orthogonal multiple access (NOMA) networks to achieve secure communication via artificial jamming, where the multi-antenna base station sends the NOMA and jamming superposition to the legitimate users with the assistance of IRS, in the presence of a single-antenna passive eavesdropper. The transmit beamforming, the jamming vector and the IRS reflecting vector are jointly optimized to maximize the sum rate of legitimate users, satisfying the quality of service requirement, the IRS reflecting constraint and the successive interference cancellation (SIC) decoding condition. In addition, the received jamming power is adapted at the highest level at all legitimate users for successful cancellation via SIC. To address it, we first decompose it into two subproblems, and then each subproblem is converted into convex using successive convex approximation. Then an alternate algorithm is proposed to solve them iteratively. Numerical results demonstrate the secure transmission in the proposed IRS-NOMA scheme can be effectively guaranteed with the assistance of artificial jamming.