Multi-User Millimeter Wave Channel Estimation Using Generalized Block OMP Algorithm

TL;DR

This paper introduces a generalized block OMP algorithm for multi-user millimeter wave channel estimation, demonstrating improved beamforming gain over traditional methods through simulations.

Contribution

It proposes a novel G.BOMP framework applicable to different multi-user channel estimation strategies in mm wave systems.

Findings

G.BOMP outperforms conventional OMP in beamforming gain

Three strategies for channel estimation are analyzed

Simulation results confirm the effectiveness of G.BOMP

Abstract

In a multi-user millimeter (mm) wave communication system, we consider the problem of estimating the channel response between the central node (base station) and each of the user equipments (UE). We propose three different strategies: 1) Each UE estimates its channel separately, 2) Base station estimates all the UEs channels jointly, and 3) Two stage process with estimation done at both UE and base station. Exploiting the low rank nature of the mm wave channels, we propose a generalized block orthogonal matching pursuit (G.BOMP) framework for channel estimation in all the three strategies. Our simulation results show that, the average beamforming gain of the G.BOMP algorithm is higher than that of the conventional OMP algorithm and other existing works on the multi-user mm wave system.