Joint Transceiver and Power Splitting Optimization for Multiuser MIMO SWIPT Under MSE QoS Constraints

• Published in: IEEE transactions on vehicular technology Vol. 66; no. 8; pp. 7123 - 7135
• Main Authors: Zhang, Haixia, Dong, Anming, Jin, Shi, Yuan, Dongfeng
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Codes; Complexity; Computer simulation; Decoding; Downlink; Energy harvesting; Energy harvesting (EH); Feasibility studies; Iterative methods; mean-squared error (MSE); MIMO; multi-input multi-output (MIMO); Optimization; Power transfer; Quality of service; quality of service (QoS); Radio equipment; Receivers; Semidefinite programming; semidefinite programming (SDP); semidefinite relaxation (SDR); simultaneous wireless information and power transfer (SWIPT); Splitting; transceiver design; Transceivers; Wireless communication
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2017.2674976

This paper studies the joint transceiver design and power splitting (JTDPS) for a downlink multiuser multiple-input multiple-output (MU-MIMO) simultaneous wireless information and power transfer (SWIPT) system. Aiming to minimize the transmit power by jointly optimizing the transmitter at a base station, the power splitting (PS) factors, and information decoding receivers at mobile stations (MSs) subject to both the predefined mean-square error (MSE) and the given energy harvesting (EH) constraints. The formulated problem is a nonconvex JTDPS optimization problem, which cannot be solved directly. To solve it, an iterative optimization framework composed of joint transmitter and PS factor optimization (JTxPS) subproblem and a receiver side minimum mean-square error (MMSE) minimization subproblem is proposed. By iteratively solve the two subproblems, the original optimization problem is solved. The sufficient and necessary conditions for the feasibility of the formulated JTDPS and the JTxPS problems are analyzed. Theoretical analysis shows that the JTxPS subproblem is also nonconvex and can be solved by using convex semidefinite programming (SDP). Considering the high computational complexity of the SDP-based scheme, a low-complexity suboptimal scheme based on the traditional MMSE transceiver design is also proposed. Simulations are done to validate the theoretical analysis, and results show the effectiveness of the proposed schemes.