MIMO-Kanalschätzung für mmWave basierend auf Deep Learning mit Außerbandinformation

Abstract

Next-generation multiple-input multiple-output (MIMO) systems will include both below 6 GHz and millimeter wave (mmWave) frequency bands in order to meet the increasing demands for high data rates. Enabling MIMO links usually requires precise channel estimation, an especially demanding task at mmWave frequencies because of the low signal-to-noise ratio (SNR). In this work, we present a novel approach for mmWave MIMO channel estimation that incorporates deep learning with sub-6 GHz out-of-band information. Our approach is based on a convolutional neural network (CNN) architecture and is compared with both deep learning approaches based entirely on in-band information and other out-of-band based approaches. The simulation results show that the proposed deep learning approach with out-of-band support outperforms existing alternative approaches in terms of mean squared channel error (MSE).