Aristotle University Of Thessaloniki (AUTH)
Artificial intelligence methods in wireless communications
Abstract
dc:descriptionThe application of artificial intelligence (AI) within wireless Communications is a transformative step forward in meeting the escalating complexity and requirements of modern network systems. This thesis investigates advanced AI and machine learning (ML) techniques designed to enhance the performance, efficiency, and adaptability of fifth-generation (5G) and beyond 5G (B5G) networks. The major areas of focus include visible light positioning (VLP), radio environment mapping (REM), and three-dimensional (3D) indoor localization, as well as accurate propagation modeling, essential for optimizing network planning and dynamic, real-time decision-making. Diverse ML models are rigorously assessed for their efficacy in addressing the sophisticated requirements posed by these wireless applications across heterogeneous and challenging environments. Aiming to improve model reliability and interpretability, this study also integrates explainable AI (XAI) methodologies. These XAI tools provide critical insights into model behavior, promoting transparency and fostering greater trust in AI-driven network solutions. Furthermore, evolutionary algorithms (EAs) are utilized for VLP systems and to optimize antenna designs, to achieve high gain and wide bandwidth for B5G and 6G applications. In parallel, deep learning (DL)-based beamforming techniques are explored, significantly enhancing connectivity and operational efficiency in high-frequency, high-mobility contexts, such as those encountered in massive multiple-input-multiple-output (MIMO) systems. These innovations enable low-complexity, scalable solutions that meet the stringent real-time demands of next-generation wireless infrastructure. Through a series of validations, the research confirms that AI-driven techniques can significantly reduce complexity and enhance the scalability of network operations, thus meeting the stringent requirements of next-generation wireless infrastructures. Furthermore, the findings elaborate on how the tailored application of various AI methodologies can be strategically deployed to address the challenges posed by next-generation wireless network requirements.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Rekkas, Vasileios-Panagiotis
- Ρέκκας, Βασίλειος-Παναγιώτης
Subjects
dc:subject × 20- Ασύρματες επικοινωνίες
- Τεχνητή νοημοσύνη
- Μηχανική μάθηση
- Επεξηγήσιμη τεχνητή νοημοσύνη
- Wireless communications
- Artificial intelligence
- Machine learning
- Explainable artificial intelligence
- Φυσική
- Φυσικές Επιστήμες
- Εφαρμοσμένη φυσική
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Physical Sciences
- Natural Sciences
- Applied Physics
- 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/58501