Joint optimization of antenna selection and beamforming in MIMO SWIPT systems with bidirectional communication

Abstract

Abstract This paper investigates joint optimization of antenna selection and beamforming in multiple‐input multiple‐output simultaneous wireless information and power transfer systems with bidirectional communication. The downlink–uplink rate region is used as the performance metric. The formulated optimization problem is non‐convex mixed integer programming, which is challenging to solve. The problem is first converted into a form of a quadratically constrained quadratic program. Given fixed receive and transmit antenna sets at the energy harvesting device, the authors apply the semidefinite relaxation to obtain a convex problem for optimal beamforming which can be solved efficiently. It is proved that the semidefinite relaxation is tight. The authors solve the relaxed problem for every possible receive and transmit antenna sets and find the optimal solution by search. Moreover, a low‐complexity method to alleviate the computational complexity of the optimal solution by deriving effective heuristic algorithms for the beamforming and antenna selection is proposed. Simulation results demonstrate that both the optimal and the low‐complexity algorithms outperform the benchmark power splitting, when the circuit power consumption is sufficiently high or the number of receive antennas at the energy harvesting device is sufficiently large. Also, a near‐optimal performance can be achieved by the proposed low‐complexity method.

Keywords

• Optimisation techniques
• Antenna arrays
• Signal processing and detection
• Radio links and equipment
• Telecommunication systems (energy utilisation)
• Wireless power transmission