Comparator Network-Aided Receivers for Spatially and Temporally Correlated MIMO Systems with 1-Bit Quantization

The low-resolution aware linear minimum mean-squared error (LRA-LMMSE) channel estimator for low-resolution MIMO receivers has demonstrated a significant reduction in mean-squared error (MSE) through the integration of a comparator network. This network consists of multiple simple comparators that produce binary outputs. In this study, we propose the Kalman filter-based channel estimator with comparator networks (KFB-CN) for temporally and spatially correlated channels for MIMO system employing 1-bit ADCs and a comparator network. After providing a mathematical derivation of the real-valued Kalman filter system and observation models, we demonstrate, through numerical simulations, that the KFB-CN outperforms the original KFB estimator. Additionally, we propose a comparator network selection algorithm which switches the comparators used on-the-fly to adapt to changes in the channel correlation coefficients.