Joint CSI Estimation, Beamforming and Scheduling Design for Wideband Massive MIMO System

TL;DR

This paper introduces a joint design for channel estimation, beamforming, and scheduling in wideband massive MIMO systems, significantly increasing the number of UEs estimated and improving performance over traditional OFDM methods.

Contribution

It presents a novel joint approach that enhances CSI estimation capacity and integrates scheduling and beamforming, with analytical validation and practical implementation simplicity.

Findings

Maximum of  UEs can be estimated with the proposed method.

The approach outperforms conventional OFDM in LTE channel simulations.

The design is linear, simple, and effective for wideband massive MIMO.

Abstract

This paper proposes a novel approach for designing channel estimation, beamforming and scheduling jointly for wideband massive multiple input multiple output (MIMO) systems. With the proposed approach, we first quantify the maximum number of user equipments (UEs) that can send pilots which may or may not be orthogonal. Specifically, when the channel has a maximum of $L$ multipath taps, and we allocate $\tilde{M}$ sub-carriers for the channel state information (CSI) estimation, a maximum of $\tilde{M}$ UEs CSI can be estimated ($L$ times compared to the conventional CSI estimation approach) in a massive MIMO regime. Then, we propose to schedule a subset of these UEs using greedy based scheduling to transmit their data on each sub-carrier with the proposed joint beamforming and scheduling design. We employ the well known maximum ratio combiner (MRC) beamforming approach in the uplink channel data transmission. All the analytical expressions are validated via numerical results, and the superiority of the proposed design over the conventional orthogonal frequency division multiplexing (OFDM) transmission approach is demonstrated using extensive numerical simulations in the long term evolution (LTE) channel environment. The proposed channel estimation and beamforming design is linear and simple to implement.