Abstract:
Over the past decade, we have witnessed an extraordinary surge in the proliferation of
connected wireless devices, the sheer volume of which has reached billions. Amidst
this rapid technological growth, another pressing concern has emerged: the energy
consumption associated with protocols for communication.
In this context, the massive MIMO (mMIMO) technology has emerged as a beacon of
promise. By equipping base sataion (BS) with an extensive array of antennas—whether
collocated or dispersed— mMIMO enables the simultaneous servicing of huge amount
of users within the specific time-frequency resource. This innovative approach aligns
perfectly with the aforementioned requirements and positions itself as a frontrunner for
driving the evolution of 5G and beyond.
In this dissertation, we delve deeply into the nuanced performance metrics of
mMIMO systems. Our exploration extends to the introduction of an innovative pilot
assignment scheme aimed at enhancing channel estimation (CE) and signal detection
accuracy. By meticulously investigating these elements, we strive to support the ongoing
development and optimization of the transformative technology, ensuring it meets the
robust demands of tomorrow’s wireless communication landscape.
In this comprehensive study, an innovative space-time transimission scheme (STTS)
is introduced that aims to surmount the inherent problems associated with channel
estimation, particularly when utilizing orthogonal pilot information in both collocated
and distributed MIMO systems equipped with numerous transmitting and receiving
antennas. The fundamental challenge arises from the necessity of acquiring accurate
channel information through orthogonal pilots, which invariably introduces pilot
overhead for channel estimation. This overhead can lead to critical bandwidth
insufficiencies, prompting a delicate trade-off between the required quantity of pilots for
effective CE and the overall spectral efficiency (SE) of the system.
The issue of data symbol detection is tackled, the MLD method, a robust approach
is employed that consistently addresses the complexities associated with parameter
estimation challenges. Moreover, singular value decomposition (SVD) is used to derive
the MGF, a mathematical tool crucial for calculating the symbol error rate (SER) using
M-ary phase shift key (M-PSK). The ergodic capacity emerges as a pivotal parameter for
ensuring reliable communication within this framework. The plot for ergodic capacity,
offering a tangible visualization of the method’s efficacy in enhancing communication
reliability is obtained.
In the analysis, the MGF of the instantaneous signal to noise (SNR) is harnessed
to develope an approximate expression for the SER in the proposed STTS framework.
Interestingly, it is discovered that the diversity order is one less than the amount of
receiver antennas utilized in this innovative scheme, highlighting an intricate relationship
iv
between these parameters.
Furthermore, an exhaustive examination of the effects of pilot sequence length on the
overall execution of the proposed transmission scheme is done, delving into the nuances
of communication theory concepts such as the probaibility density function (PDF) and
cumulcative distribution function (CDF). Throughrigorous simulations, PDF and CDF
plots across various degrees of freedom and system configurations are obtained. The
findings reveal that as the degree of freedom improves, the suitably normalized sum of
the channel information exhibits a tendency to converge towards a normal distribution
within the STTS framework.
The architecture of our transmitter is notably sophisticated, featuring a substantial
array of antennas specifically designed to accommodate multiple users. Each user is
assigned a distinct group of antennas, and the transmitter adeptly employs tailored
beamforming vectors for every user group to broadcast signals. On the receiving end,
unique combining vectors are implemented for accurate signal detection, ensuring
optimal performance.
To effectively nullify interference from the signals of unintended users, our proposed
scheme capitalizes on the concept of a null space, an advanced technique that enhances
signal clarity and reliability. The computation of the combining vector is anchored in
the maximum eigenvalue criterion, an established method that augments the scheme’s
effectiveness.
The extensive simulations and analyses unequivocally demonstrate that the proposed
STTS exhibits significant improvements in performance when a higher amount of
antennas are implemented at either the transmitter (Tx) or the user end, aptly showcasing
the potential of this advanced transmission scheme in modern communication systems.
The STTS, combined with DNN, play a crucial role in estimating communication
channels in OFDM systems. This approach allows for more accurate and efficient channel
estimation, which is essential for improving signal quality and minimizing interference
in wireless communications. By leveraging the powerful pattern recognition capabilities
of DNNs, the STTC can enhance the performance of OFDM systems, ensuring reliable
data transmission even in challenging environments.