Adaptive Learning-Based Detection for One-Bit Quantized Massive MIMO Systems

TL;DR

This paper introduces an adaptive learning-based detection method for one-bit quantized massive MIMO systems that improves performance by using dithering noise and adaptive probability estimation, avoiding explicit channel estimation.

Contribution

It proposes an adaptive dithering-and-learning framework that enhances detection accuracy in one-bit massive MIMO without needing channel estimation.

Findings

Outperforms previous fixed-dithering methods

Achieves detection performance close to ML with perfect channel knowledge

Effective across various SNR regimes

Abstract

We propose an adaptive learning-based framework for uplink massive multiple-input multiple-output (MIMO) systems with one-bit analog-to-digital converters. Learning-based detection does not need to estimate channels, which overcomes a key drawback in one-bit quantized systems. During training, learning-based detection suffers at high signal-to-noise ratio (SNR) because observations will be biased to +1 or -1 which leads to many zero-valued empirical likelihood functions. At low SNR, observations vary frequently in value but the high noise power makes capturing the effect of the channel difficult. To address these drawbacks, we propose an adaptive dithering-and-learning method. During training, received values are mixed with dithering noise whose statistics are known to the base station, and the dithering noise power is updated for each antenna element depending on the observed pattern of the output. We then use the refined probabilities in the one-bit maximum likelihood detection rule. Simulation results validate the detection performance of the proposed method vs. our previous method using fixed dithering noise power as well as zero-forcing and optimal ML detection both of which assume perfect channel knowledge.