Joint Pilot Allocation and Power Control to Enhance Max-Min Spectral Efficiency in TDD Massive MIMO Systems

Honggui Deng,Gang Liu,Gaofeng Peng,Wenhui Wang,Xuewen Qian

doi:10.1109/access.2019.2947113

Honggui Deng, Gang Liu + Show 3 more

Open Access

https://doi.org/10.1109/access.2019.2947113

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2019
Citations: 8	License type: CC BY 4.0

Affiliation: Central South University, Changsha Normal University

Abstract

In time division duplex (TDD) massive multiple-input multiple-output (MIMO) systems, pilot contamination seriously limits the improvement of spectral efficiency (SE). In order to provide good communication quality for every user, this paper devotes to enhance minimum spectral efficiency of the system. Firstly, a multi-cell multi-user system model is established to derive the objective function of Max-Min spectral efficiency. Through analysis of the mathematical expressions, we found that it is not a convex optimization problem for joint pilot allocation and power control, but a non-deterministic polynomial (NP) problem that is difficult to deal with. In order to solve the problem, a new algorithm is proposed to decompose the problem into two easy-to-handle sub-problems. We employ weight graph coloring (WGC) algorithm to deal with the sub-problem of pilot allocation, where uplink transmission power is fixed. The geometric programming (GP) approach is utilized to solve the sub-problem of power control, in which the large-scale fading factor is fixed. Then the computational complexity of the proposed algorithm is analyzed. Compared with the ideal optimal solution algorithm that requires exhaustive search, our algorithm has much less computational complexity. Finally, the simulation results demonstrate that the proposed algorithm significantly improves the minimum spectral efficiency, which is close to the ideal optimal solution.

Highlights

Massive multiple-input multiple-output (MIMO) has become one of the key technologies of the fifth-generation networks (5G) due to the significant improvements on the performance of spectral efficiency, energy efficiency and link reliability [1]–[4]
D) Random pilot assignment (RPA)-PPC: The algorithm only optimizes the pilot power according to the power control optimization method described in subsection B of Section III
Compared with RPA-UPC, the weight graph coloring (WGC)-PD-UPC and ESPA-UPC has increased 2.372b/s/Hz and 2.111b/s/Hz in terms of average minimum spectral efficiency respectively. These results indicate that pilot allocation, pilot power and uplink data power control should be optimized to maximize the minimum spectral efficiency

Summary

Introduction

Massive MIMO has become one of the key technologies of the fifth-generation networks (5G) due to the significant improvements on the performance of spectral efficiency, energy efficiency and link reliability [1]–[4]. The TDD mode is generally utilized in massive MIMO systems since it can facilitate channel estimation by the reciprocity of uplink and downlink channels in coherence time [5]. In TDD massive MIMO system, the reuse of orthogonal pilots is adopted to solve the problem of limited pilot resource. This strategy leads to interference among users in different cells using the same pilot, which is referred to as pilot contamination. Pilot contamination seriously affects the accuracy of channel estimation and transmission rate, which is recognized as the performance bottleneck of massive MIMO systems

Objectives

Results

Conclusion