Machine Learning Detectors for MU-MIMO Systems With One-Bit ADCs

Abstract

We consider an uplink multiuser multiple-input multiple-output (MU-MIMO) system with one-bit analog-to-digital converters (ADCs). In this system, the construction of a low-complexity detector is quite challenging due to the non-linearity of an end-to-end channel transfer function. Recently, a supervisedlearning (SL) detector was proposed by modeling the complex non-linear function as a tractable Bernoullimixture model. It achieves an optimal maximum-likelihood (ML) performance, provided the channel state information (CSI) is perfectly known at a receiver. However, when a system-size is large, SL detector is not practical because of requiring a large amount of labeled data (i.e., pilot signals) to estimate the model parameters. We address this problem by proposing a semi-supervised learning (SSL) detector in which both pilot signals (i.e., labeled data) and some part of data signals (i.e., unlabeled data) are used to estimate them via expectation-maximization (EM) algorithm. We further extend the proposed detector for timevarying channels, by leveraging the idea of online learning, which is called online-learning (OL) detector. Simulation results demonstrate that the proposed SSL detector can achieve the almost same performance of the corresponding SL detector with significantly lower pilot overhead. In addition, it is shown that the proposed OL detector is more robust to channel variations compared with the existing detectors.

Massive MIMO Systems for 5G and beyond Networks—Overview, Recent Trends, Challenges, and Future Research Direction

“…The conventional signal detection methods are computationally very complex and inefficient for large antennas systems like massive MIMO. Several semi-supervised learning (SSL) [162] and supervised learning (SL) [163] approach have been proposed and provide more robust performance. Several other uses of machine learning and deep learning have been presented in Reference [164][165][166][167].…”

“…The use of machine learning and deep learning algorithms during massive MIMO channel estimation to predict statistical channel characteristics is an exciting area of research. Several experiments have been conducted recently to explore machine learning and deep learning for massive MIMO channel estimation, user scheduling, beamforming, and signal detection [90,[149][150][151][152][153][154][155][156][157][158][159][160][161][162][163][164][165][166][167].…”

Deep Learning for Estimation and Pilot Signal Design in Few-Bit Massive MIMO Systems

“…Let P B (x) be the objective function of (27). Note that P B (x) is a quadratic function of x and thus convex.…”

“…However, the nML method is non-robust at high signal-to-noise ratios (SNRs) when the channel state information (CSI) is not perfectly known. The learning-based method in [27] is a blind detection method for which CSI is not required, but it is only applicable to MIMO systems with a small number of transmit antennas and only low-dimensional constellations. Various one-bit linear detectors were introduced in [28], [29].…”

“…However, the resulting methods can be sophisticated and expensive to implement. Unlike the blind detection method in [27] which was developed for one-bit systems, the learning-based blind detection methods in [38], [39] are applicable for few-bit systems, but they are also restricted to MIMO systems with a small number of transmit antennas and only low-dimensional constellations. The BMMSE and BWZF detection methods in [22] are linear detectors and thus simple and applicable for large-scale MIMO systems.…”