Non-Orthogonal Multiple Access With Improper Gaussian Signaling

• Published in: IEEE journal of selected topics in signal processing 2019-06, Vol.13 (3), p.496-507
• Main Authors: Tuan, Hoang Duong, Nasir, Ali Arshad, Nguyen, Ha H., Duong, Trung Q., Poor, H. Vincent
• Format: Article
• Language: English
• Subjects: <inline-formula xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <tex-math notation="LaTeX"> 2\times 2</tex-math> </inline-formula> linear matrix inequality; Algorithms; Array signal processing; Beamforming; Covariance matrices; improper Gaussian signaling (IGS); Interference; Linear matrix inequalities; Matrix methods; multi-cell networks; NOMA; non-orthogonal multiple access (NOMA); nonconvex optimization; Nonorthogonal multiple access; Optimization; Signal processing; Signaling; Throughput; Transmission beamforming; Wireless communications; Wireless networks
• ISSN: 1932-4553; 1941-0484
• DOI: 10.1109/JSTSP.2019.2901993

Improper Gaussian signaling (IGS) helps to improve the throughput of a wireless communication network by taking advantage of the additional degrees of freedom in signal processing at the transmitter. This paper exploits IGS in a general multiuser multi-cell network, which is subject to both intra-cell and inter-cell interference. With IGS under orthogonal multiple access (OMA) or non-orthogonal multiple access (NOMA), designs of transmit beamforming to maximize the users' minimum throughput subject to transmit power constraints are addressed. Such designs are mathematically formulated as nonconvex optimization problems of structured matrix variables, which cannot be solved by popular techniques such as weighted minimum mean square error or convex relaxation. By exploiting the lowest computational complexity of 2\times 2 linear matrix inequalities, lower concave approximations are developed for throughput functions, which are the main ingredients for devising efficient algorithms for finding solution of these difficult optimization problems. Numerical results obtained under practical scenarios reveal that there is an almost two-fold gain in the throughput by employing IGS instead of the conventional proper Gaussian signaling under both OMA and NOMA; and NOMA-IGS offers better throughput compared to that achieved by OMA-IGS.