Abstract
This paper presents a novel channel estimation technique for the multi-user massive multiple-input multiple-output (MU-mMIMO) systems using angular-based hybrid precoding (AB-HP). The proposed channel estimation technique generates group-wise channel state information (CSI) of user terminal (UT) zones in the service area by deep neural networks (DNN) and fuzzy c-Means (FCM) clustering. The slow time-varying CSI between the base station (BS) and feasible UT locations in the service area is calculated from the geospatial data by offline ray tracing and a DNN-based path estimation model associated with the 1-dimensional convolutional neural network (1D-CNN) and regression tree ensembles. Then, the UT-level CSI of all feasible locations is grouped into clusters by a proposed FCM clustering. Finally, the service area is divided into a number of non-overlapping UT zones. Each UT zone is characterized by a corresponding set of clusters named as UT-group CSI, which is utilized in the analog RF beamformer design of AB-HP to reduce the required large online CSI overhead in the MU-mMIMO systems. Then, the reduced-size online CSI is employed in the baseband (BB) precoder of AB-HP. Simulations are conducted in the indoor scenario at 28 GHz and tested in an AB-HP MU-mMIMO system with a uniform rectangular array (URA) having 16x16=256 antennas and 22 RF chains. Illustrative results indicate that 91.4% online CSI can be reduced by using the proposed offline channel estimation technique as compared to the conventional online channel sounding. The proposed DNN-based path estimation technique produces same amount of UT-level CSI with runtime reduced by 65.8% as compared to the computationally expensive ray tracing.
Highlights
The generation communication networks experience the severe challenge of an increasing number of connections between the base station (BS) and the dense-deployed user terminals (UTs)
We study and develop deep neural networks (DNN)-based path estimation models, including a one-step model based on the feedforward neural network (FFNN) and three two-step models based on the 1-dimensional convolutional neural network (1D-CNN), via three steps: (i) extract input features from the 3D geospatial data by data preprocessing, (ii) learn a small number of true paths of selected feasible UT locations and corresponding input features in the training set by a DNN-based model, (iii) predict all paths for the remaining feasible UT locations only according to geospatial data by the trained DNN-based model
ILLUSTRATIVE RESULTS To evaluate the performance of the proposed DNN-based path estimation framework, we present a one-step model trained with 70% of paths and three two-step models trained with 70%, 50% and 30% of paths required for the channel estimation for the indoor 28 GHz scenario
Summary
The generation communication networks experience the severe challenge of an increasing number of connections between the base station (BS) and the dense-deployed user terminals (UTs). Obtaining the slow time-varying CSI between the BS and UT groups without any knowledge of the online channel, especially estimating the angle information, is of great importance to fully drop the large-size CSI overhead required for the RF beamformer. B. CONTRIBUTIONS AND ORGANIZATION In this paper, a learning-based channel estimation approach for the MU-mMIMO hybrid precoding RF beamformer has been presented. The UT-level CSI produced by offline ray tracing and the DNN-based path estimation technique forms cluster-level CSI by a proposed FCM clustering algorithm. This paper introduces a deep learning model for the UT-level CSI acquisition, which generates same amount of slow time-varying UT-level CSI for feasible UT locations in the service area with runtime significantly reduced. MASSIVE-MIMO HYBRID PRECODING & GEOMETRY-BASED 3D CHANNEL MODEL we present the hybrid precoding architecture and the geometry-based 3D channel model
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
