Joint beamforming and power-splitting optimization for SWIPT-enabled MISO full-duplex two-way cooperative NOMA systems

Abstract

In this paper, a multiple-input single-output (MISO) full-duplex two-way cooperative non-orthogonal multiple access (NOMA) system is considered. The relay is energy-restricted and may not accomplish the information exchange between two end users. To address this problem, simultaneous wireless information and power transfer (SWIPT) is proposed, which employs power splitting (PS) at the relay to harvest energy. To maximize the harvested energy of the relay under QoS and power constraints, a joint optimization problem of the beamforming vectors and the PS ratio is formulated. The problem is nonconvex, and the original problem is transformed into a tractable form by applying the semidefinite relaxation (SDR) technique. Then a bilevel programming (BLP) based method is proposed to obtain an approximately optimal solution. Moreover, to reduce the computational complexity, a successive convex approximation (SCA)-based iterative algorithm is developed to derive a suboptimal solution. Numerical results show that our proposed strategy significantly enhances the energy harvesting performance compared with the existing transmission strategies.