Massive MIMO Channel Prediction: Kalman Filtering vs. Machine Learning

TL;DR

This paper compares Kalman filter and machine learning methods for predicting massive MIMO channels using realistic 3GPP SCM channels, highlighting their accuracy, complexity, and practical applicability.

Contribution

It develops and evaluates a VKF and an ML-based channel predictor using realistic SCM channels, introducing a low-complexity mobility estimator for massive MIMO.

Findings

Both predictors outperform outdated channels in accuracy and data rate.

ML predictor has higher initial complexity but lower operational complexity after training.

VKF predictor exploits AR parameters estimated from SCM channels.

Abstract

This paper focuses on channel prediction techniques for massive multiple-input multiple-output (MIMO) systems. Previous channel predictors are based on theoretical channel models, which would be deviated from realistic channels. In this paper, we develop and compare a vector Kalman filter (VKF)-based channel predictor and a machine learning (ML)-based channel predictor using the realistic channels from the spatial channel model (SCM), which has been adopted in the 3GPP standard for years. First, we propose a low-complexity mobility estimator based on the spatial average using a large number of antennas in massive MIMO. The mobility estimate can be used to determine the complexity order of developed predictors. The VKF-based channel predictor developed in this paper exploits the autoregressive (AR) parameters estimated from the SCM channels based on the Yule-Walker equations. Then, the ML-based channel predictor using the linear minimum mean square error (LMMSE)-based noise pre-processed data is developed. Numerical results reveal that both channel predictors have substantial gain over the outdated channel in terms of the channel prediction accuracy and data rate. The ML-based predictor has larger overall computational complexity than the VKF-based predictor, but once trained, the operational complexity of ML-based predictor becomes smaller than that of VKF-based predictor.