Abstract:
The growth of network devices and services has caused a substantial rise in global
data tra c demand. Cellular networks face challenges in meeting high data demands
of future wireless networks due to uneven data rates between cell centers and edges.
Cell-free massive multiple-input multiple-output (CF mMIMO) has emerged as a
promising solution to support rising data demands by leveraging distributed access
points (APs) to cooperatively serve users across a coverage area. Inter-user interference
(IUI) from unpredictable phase shifts and ine cient power allocation should be
addressed to improve performance. In this thesis, we analyzed CF mMIMO systems
using power-domain non-orthogonal multiple access (PD-NOMA) and successive interference
cancellation (SIC) with phase shift-aware channel estimation. Phase of
the line-of-site (LoS) components are modeled as a uniformly distributed random
variables to take phase-shifts concerning user equipment (UE) mobility in the estimation
process. Three estimation methods|phase-aware minimum mean square
error (PA-MMSE), non-phase-aware minimum mean square error (NPA-MMSE),
and least squares (LS)|are applied.
The favorable phase range of the APs' LoS component is investigated. For NOMA,
we employed an e cient power allocation method where UEs are sorted based on the
e ective channel gain connected to each AP. Lastly, phase-synchronized AP cooperation
is compared to unsynchronized to analyze the e ect on phase shifts. Downlink
spectral e ciency (SE) and energy e ciency (EE) are analyzed through Monte Carlo
simulations. The results demonstrate improvements in system performance, showcasing
the potential of phase shift-aware channel estimation in CF mMIMO-NOMA
systems. PA-MMSE estimator with NOMA improves the SE by 14.2% and 4.1%
from PA-MMSE and NPA-MMSE NOMA, respectively, while EE improves by 44%
and 11.1% respectively. The employed power allocation technique outperforms the
conventional method by achieving a gain of over 25.2% in SE performance.
Keywords: Cell-Free Massive MIMO, Non-Orthogonal Multiple Access, Phase Shifts,
Power Allocation, Successive Interference Cancellation, Spectral Effciency.