University of the Aegean
AI-powered cooperative integration of aerial and terrestrial networks for enhanced end-user connectivity
Abstract
dc:descriptionThe evolution of wireless communications toward B5G and 6G systems introduces unprecedented requirements in terms of capacity, coverage, reliability, and adaptability. Conventional terrestrial infrastructures alone are increasingly unable to satisfy these demands, particularly in scenarios characterized by dynamic user distributions, sudden traffic surges, infrastructure failures, and heightened security threats. As a result, the integration of aerial communication platforms with ground networks has emerged as a promising approach for enabling flexible, resilient, and high-performance wireless ecosystems. Unmanned Aerial Vehicles (UAVs) possess distinctive characteristics such as rapid deployability, mobility, and on-demand coverage, allowing them to effectively complement terrestrial infrastructures. Nevertheless, their incorporation into communication networks raises important challenges related to deployment optimization, cooperative transmission, resource allocation, and secure communication. Addressing these challenges, this thesis develops Artificial Intelligence (AI)- and Machine Learning (ML)-driven frameworks that facilitate efficient and secure cooperation between aerial and terrestrial network components, supporting the performance objectives of next-generation wireless systems. Chapter 1 provides the necessary background on UAV-enabled communications and examines the technological foundations that enable intelligent collaboration between aerial and terrestrial networks. It further defines the research motivation and objectives, highlighting the importance of AI-driven methodologies for addressing issues such as UAV placement, resource management, cooperative networking, and physical layer security. Chapter 2 proposes an intelligent integration framework for aerial–terrestrial cooperative networks in which UAVs are dynamically deployed through clustering-based techniques to ensure seamless connectivity. The presented unsupervised learning approach determines suitable UAV placement for traffic offloading and emergency communication scenarios, thereby improving service continuity in dense environments and areas affected by infrastructure disruptions through Coordinated Multi-Point (CoMP) transmission. Chapter 3 investigates the security of aerial–terrestrial networks operating in the presence of potentially untrusted relays. An AI-driven framework is introduced to optimize the three-dimensional placement of UAV relays, power allocation strategies, and antenna selection while incorporating cooperative jamming mechanisms. By combining clustering algorithms, Particle Swarm Optimization (PSO), and deep neural network-based decision processes, the proposed solutions achieve notable improvements in secrecy rate and reductions in secrecy outage probability, strengthening the robustness of future wireless infrastructures. Chapter 4 extends the cooperative aerial–terrestrial networking paradigm by introducing a UAV–Device-to-Device (D2D) cooperative communication framework that integrates UAV-assisted Non-Orthogonal Multiple Access (NOMA) with D2D communications. By leveraging both supervised and unsupervised learning techniques, the proposed framework optimizes UAV placement, D2D user pairing, and resource allocation to enable efficient spectrum sharing. Simulation results demonstrate that the proposed UAV–D2D cooperative scheme achieves significant improvements in spectral efficiency and overall system throughput, while guaranteeing that each user attains a higher data rate than that achievable under conventional Orthogonal Multiple Access (OMA) schemes. Finally, the thesis summarizes its primary findings and outlines future research directions, including scalable multi-UAV coordination, integration with non-terrestrial network elements, and the development of energy-aware optimization strategies for sustainable aerial communication systems. Overall, the results presented in this thesis indicate that combining AI-driven decision-making with UAV-enabled networking can substantially enhance connectivity, security, and resource efficiency.
Degree
thesis:*- Grantor dc:publisher
- University of the Aegean
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tsipi, Lefter
- Τσίπη, Λευτέρ
Subjects
dc:subject × 12- Τεχνητή νοημοσύνη
- Εναέρια-Επίγεια Δίκτυα
- Μη επανδρωμένα αεροχήματα
- Aerial–Terrestrial Networks
- Artificial intelligence
- UAV Communications
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Communication engineering and systems, Telecommunications
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/61261