Improving Energy Efficiency of Massive MIMO Relay Systems using Power Bisection Allocation for Cell-Edge Users

TL;DR

This paper enhances the energy efficiency of massive MIMO relay systems by proposing power bisection algorithms for optimal power allocation at base stations and relays, especially benefiting cell-edge users.

Contribution

It introduces a novel PBODS algorithm combining power bisection and exhaustive search, and an alternative AOP method for reduced complexity with minimal performance loss.

Findings

PBODS approaches optimal performance with lower complexity

AOP offers a practical trade-off between complexity and efficiency

Methods improve energy efficiency for cell-edge users in massive MIMO systems

Abstract

This paper studies downlink energy efficiency (EE) of a cellular massive Multiple Input Multiple Output (MIMO) system with massive number of antennas at both base station (BS) and relay considering cell-edge users which is less investigated in the literature. Some improving methods for base station (BS) and relay power allocation under quality of service constraints are proposed. The EE optimization problem is provided over BS and relay power variables (power dimensions). Considering the quasi-concavity property of the EE function relative to each power variable, we propose to use a power bisection algorithm (PBA) in one dimension, followed by the exhaustive search in the other dimension (ODS algorithm) which is called PBODS algorithm and also a method to limit the range of power variables. The results show that the performance of the PBODS algorithm approaches the performance of optimal solution which is exhaustive search in both dimensions while it has a lower complexity. In addition to that we suggest to use the PBA in two dimensions alternatively as the sub-optimal alternative optimization (AOP) algorithm. The complexity can highly be reduced using the AOP algorithm with a slight but acceptable degradation in performance which makes the algorithm much suitable for practical use.