A Reduced Complexity Ungerboeck Receiver for Quantized Wideband Massive SC-MIMO

TL;DR

This paper introduces a low-complexity iterative receiver for quantized wideband SC-MIMO systems that improves performance without relying on channel sparsity, using Bussgang decomposition and Ungerboeck factorization.

Contribution

It proposes a novel message passing based receiver with reduced complexity and no need for a whitening filter, along with a channel estimator that avoids cyclic-prefix overhead.

Findings

Significant performance gains over existing receivers.

Complexity reduction compared to prior methods.

Effective channel estimation without cyclic-prefix.

Abstract

Employing low resolution analog-to-digital converters in massive multiple-input multiple-output (MIMO) has many advantages in terms of total power consumption, cost and feasibility of such systems. However, such advantages come together with significant challenges in channel estimation and data detection due to the severe quantization noise present. In this study, we propose a novel iterative receiver for quantized uplink single carrier MIMO (SC-MIMO) utilizing an efficient message passing algorithm based on the Bussgang decomposition and Ungerboeck factorization, which avoids the use of a complex whitening filter. A reduced state sequence estimator with bidirectional decision feedback is also derived, achieving remarkable complexity reduction compared to the existing receivers for quantized SC-MIMO in the literature, without any requirement on the sparsity of the transmission channel. Moreover, the linear minimum mean-square-error (LMMSE) channel estimator for SC-MIMO under frequency-selective channel, which do not require any cyclic-prefix overhead, is also derived. We observe that the proposed receiver has significant performance gains with respect to the existing receivers in the literature under imperfect channel state information.