Beamspace Aware Adaptive Channel Estimation for Single-Carrier Time-varying Massive MIMO Channels

TL;DR

This paper introduces a novel beamspace aware adaptive channel estimation method for time-varying massive MIMO systems, improving accuracy and robustness while reducing pilot overhead and hardware complexity.

Contribution

It proposes a new statistical pre-beamformer design algorithm for sequential beam construction in time-varying MIMO channels using reduced rank Kalman filtering.

Findings

Achieves near-optimal beam pattern design with low complexity.

Demonstrates robustness to pilot interference.

Reduces pilot overhead and hardware complexity.

Abstract

In this paper, the problem of sequential beam construction and adaptive channel estimation based on reduced rank (RR) Kalman filtering for frequency-selective massive multiple-input multiple-output (MIMO) systems employing single-carrier (SC) in time division duplex (TDD) mode are considered. In two-stage beamforming, a new algorithm for statistical pre-beamformer design is proposed for spatially correlated time-varying wideband MIMO channels under the assumption that the channel is a stationary Gauss-Markov random process. The proposed algorithm yields a nearly optimal pre-beamformer whose beam pattern is designed sequentially with low complexity by taking the user-grouping into account, and exploiting the properties of Kalman filtering and associated prediction error covariance matrices. The resulting design, based on the second order statistical properties of the channel, generates beamspace on which the RR Kalman estimator can be realized as accurately as possible. It is observed that the adaptive channel estimation technique together with the proposed sequential beamspace construction shows remarkable robustness to the pilot interference. This comes with significant reduction in both pilot overhead and dimension of the pre-beamformer lowering both hardware complexity and power consumption.