A Minorization-Maximization Method for Optimizing Sum Rate in Non-Orthogonal Multiple Access Systems

TL;DR

This paper introduces a minorization-maximization algorithm to optimize sum rate in MISO NOMA systems, demonstrating improved spectral and power efficiency over traditional methods with manageable computational complexity.

Contribution

It develops a novel MMA-based approach for non-convex sum rate maximization in NOMA, including conditions for problem simplification and efficiency analysis.

Findings

NOMA outperforms conventional multiple access in spectral efficiency.

The proposed MMA algorithm converges in a few iterations.

NOMA achieves better power efficiency with polynomial-time complexity.

Abstract

Non-orthogonal multiple access (NOMA) systems have the potential to deliver higher system throughput, compared to contemporary orthogonal multiple access techniques. For a linearly precoded multiple-input multiple-output (MISO) system, we study the downlink sum rate maximization problem, when the NOMA principles are applied. Being a non-convex and intractable optimization problem,we resort to approximate it with a minorization-maximization algorithm (MMA), which is a widely used tool in statistics. In each step of the MMA, we solve a second-order cone program, such that the feasibility set in each step contains that of the previous one, and is always guaranteed to be a subset of the feasibility set of the original problem. It should be noted that the algorithm takes a few iterations to converge. Furthermore, we study the conditions under which the achievable rates maximization can be further simplified to a low complexity design problem, and we compute the probability of occurrence of this event. Numerical examples are conducted to show a comparison of the proposed approach against conventional multiple access systems. NOMA is reported to provide better spectral and power efficiency with a polynomial time computational complexity.