Semi-Blind Strategies for MMSE Channel Estimation Utilizing Generative Priors

This paper investigates semi-blind channel estimation for massive multiple-input multiple-output (MIMO) systems. To this end, we first estimate a subspace based on all received symbols (pilot and payload) to provide additional information for subsequent channel estimation. We show how this additional information enhances minimum mean square error (MMSE) channel estimation. Two variants of the linear MMSE (LMMSE) estimator are formulated, where the first one solves the estimation within the subspace, and the second one uses a subspace projection as a preprocessing step. Theoretical derivations show the superior estimation performance of the latter method in terms of mean square error for uncorrelated Rayleigh fading. Subsequently, we introduce parameterizations of this semi-blind LMMSE estimator based on two different conditional Gaussian latent models, i.e., the Gaussian mixture model and the variational autoencoder. Both models learn the underlying channel distribution of the propagation environment based on training data and serve as generative priors for semi-blind channel estimation. Extensive simulations for real-world measurement data and spatial channel models show the superior performance of the proposed methods compared to state-of-the-art semi-blind channel estimators with respect to the MSE.