High Dimensional Covariance Matrix Estimators for Sounding Reference Signals based Channel Estimates in massive MIMO systems
Loading...
URL
Journal Title
Journal ISSN
Volume Title
Sähkötekniikan korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
Department
Major/Subject
Mcode
ELEC3029
Language
en
Pages
66
Series
Abstract
Wireless communications and methods in signal processing have taken giant leaps of advancement in the recent era. 5G New Radio enormously expands the capabilities of mobile networks. The new cellular network technology is about the integration of wireless access technologies into one seamless experience. 5G NR’s enabling technologies such as massive MIMO, ultra-reliable low latency (URLLC), massive machine-type communications (MTC), enhanced mobile broadband connectivity (eMBB) will potentially impact many industries. Massive MIMO in 5G enable the telecommunications industry to deploy base stations equipped with features such as beamforming for enhanced user data rate in the downlink and also improve the performance in the uplink. This thesis examines high dimensional covariance matrix estimators (HDCM) (well known as \textit{shrinkage estimators}) for sounding reference signals (SRS) based beamforming in massive MIMO systems of 5G NR. SRS based eigenbeamforming corresponds to the downlink signal processing technique done in the physical layer in the base station of the cellular network. It is done using the user covariance matrix hence this thesis examines HDCM estimators for this purpose. This thesis also implements a minimum variance distortionless response (MVDR) beamformer to assess the performance of various covariance matrix estimators considered with SINR as a performance metric. The results indicate that the implemented HDCM estimators for SRS based beamforming in 5G NR do not show significant improvement in the performance compared to the sample covariance matrix (SCM). This is because the covariance is not subjected to any inversion. On the contrary, the MVDR beamformer showed significant improvement by using HDCM estimators against the SCM based approach. The analysis paves the way to consider HDCM estimators for 5G and beyond technologies as the number of antennas at base station increases and also to enable robust adaptive beamforming for cellular-based wireless communications, but it needs further research and change in architecture.Description
Supervisor
Ollila, EsaThesis advisor
Hassinen, MarkoMedeiros, Luiz