Massive MIMO Channel Estimation for Millimeter Wave Systems via Matrix Completion

TL;DR

This paper introduces a new iterative channel estimation algorithm for mmWave massive MIMO systems that jointly exploits channel sparsity and low rank properties, enabling more accurate recovery with fewer training symbols.

Contribution

A novel ADMM-based algorithm that jointly leverages channel sparsity and low rank properties for improved mmWave MIMO channel estimation with shorter training periods.

Findings

Achieves more accurate channel recovery with fewer training symbols.

Provides a globally optimal solution with fast convergence.

Outperforms existing methods in simulation tests.

Abstract

Millimeter Wave (mmWave) massive Multiple Input Multiple Output (MIMO) systems realizing directive beamforming require reliable estimation of the wireless propagation channel. However, mmWave channels are characterized by high variability that severely challenges their recovery over short training periods. Current channel estimation techniques exploit either the channel sparsity in the beamspace domain or its low rank property in the antenna domain, nevertheless, they still require large numbers of training symbols for satisfactory performance. In this paper, we present a novel channel estimation algorithm that jointly exploits the latter two properties of mmWave channels to provide more accurate recovery, especially for shorter training intervals. The proposed iterative algorithm is based on the Alternating Direction Method of Multipliers (ADMM) and provides the global optimum solution to the considered convex mmWave channel estimation problem with fast convergence properties.