AdaBoost-Based Channel Estimation in One-Bit Millimeter-Wave MIMO

Abstract

Leveraging one-bit analog-to-digital converter (ADC) instead of high resolution ADC has been introduced as a promising solution for reducing the power consumption and hardware cost of massive millimeter-wave (mmWave) multiple-input multiple-output (MIMO) systems. However, performance loss caused by discarding the amplitude information by one-bit quantizers is a significant impairment which calls for the development of more accurate channel estimators. To address this problem, a one-bit mmWave MIMO channel estimation method based on adaptive boosting (AdaBoost) which employs two-stage weak classifiers is developed. In the first stage of each weak classifier, an approximate Gaussian discriminant analysis (GDA) binary classifier is used. To capture the sparsity of mmWave channels in the angular domain, a hard thresholding operator is employed in the second stage of each weak classifier. Numerical simulations are included to demonstrate the efficiency and accuracy of the proposed channel estimator.