Performance evaluation of terrestrial and aerial communication networks
Αξιολόγηση της επίδοσης επίγειων και εναέριων δικτύων επικοινωνιών
View/
Abstract
Moving with tremendous growth toward the sixth-generation (6G) networks, it becomes more and
more clear that a key feature of the new era will be the uninterrupted connectivity to both users and
machine-type devices for supporting a huge variety of applications. To this end, the research and
industry community turns to key enabling technologies such as the millimeter-wave (mmWave)
and the unmanned aerial vehicles (UAVs). However, the introduction of these technologies comes
with several challenges. To investigate these challenges and propose solutions for the design of
6G cellular networks, detailed system-level analysis is required, which is the main objective of this
dissertation.
First, a comprehensive performance analysis is conducted for terrestrial mmWave networks, where
the user equipments (UEs) and the base stations (BSs) are equipped with antenna arrays that are
able to form directional beams. Nevertheless, with the introduction of beamforming capabilities
and the inevitable misalignment error between the two communications antennas, it becomes clear
that both polar coordinates should now be jointly considered in the user association policy and the
performance evaluation of the network. By jointly considering both polar coordinates, the probability
density function (PDF) of the maximum received power and the Laplace transform of the
aggregate interference power distributions, are derived in exact form. Subsequently, coverage probability
is examined by exploiting fundamental tools of stochastic geometry. Going one step further,
the analysis is also conducted for two other association schemes, i.e., an ideal baseline scenario and
a purely angular distance-based association scheme. The effect of a single dominant interferer in
coverage probability is also investigated for each policy scheme. As a baseline approach, a finite
mmWave network under perfect alignment between the serving BS and the receiver is also considered.
To elaborate more on system-level insights, three key interference scenarios are presented
and the network’s coverage performance is subsequently investigated.
Continuing the analysis on 6G key enabling technologies, the research interest now turns on the
integration of UAVs in cellular networks. Unfortunately, current cellular architectures have not primarily
been designed for integrating aerial nodes. Triggered by the aforementioned, the binomial
3D cellular network is proposed as a candidate cellular architecture for beyond fifth-generation
(5G) UAV networks. The binomial 3D cellular network integrates merely UAV-BSs and UAV-UEs
and it builds on the truncated octahedron cell-shape which is the closest approximation of a sphere
which can fully tessellate the three-dimensional (3D) space with no overlaps or gaps. Due to the peculiar
geometry of the truncated octahedron, conducting a performance evaluation at system-level
is particularly challenging. A key challenge is the evaluation of the volume of the sub-regions of
a truncated octahedron cell-shape in which a UAV-UE may fall. Accordingly, by exploiting the
3D binomial point process (BPP) for modeling the spatial locations on the UAV-UEs, performance
analysis in terms of coverage probability is conducted under i) intra-cell interference only, and ii)
both intra- and a worst-case inter-cell interference scenario. For the sake of completeness, coverage
probability analysis is also conducted for a finite cell modeled as a sphere which can clearly
capture the effect of the 3D coverage space while on the same time retains analytical tractability.
Accordingly, the analysis focuses to the performance under a dominant interferer approach in a
finite cell modeled as a sphere. Closed-form expressions are obtained for the coverage probability,
BER, average output SIR, and ergodic capacity. Finally, we bring more light to the debate between
cellular networks which are based on the sphere cell-shape and networks which are based on the
truncated octahedron cell-shape. Accordingly, the interest turns to the spatial deployment of the
UAV-BS and therefore a system-level performance comparison is conducted between the truncated
octahedron-based cellular network and a 3D binomial-Voronoi network in a sphere. While the
actual UAV-BS 3D deployment is neither absolutely well-defined nor totally random in practice,
performance comparison is conducted to highlight the benefits of each model as a candidate 3D
cellular UAV network for beyond-5G wireless communication systems.
Department
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
155Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα