University of Thessaly (UTH)
AI-Driven orchestration and management of cloud-native 6G networks
Abstract
dc:descriptionThe sixth generation of mobile networks is set to redefine the boundaries of wireless connectivity, enabling applications that range from holographic telepresence and pervasive AI-driven services to seamless integration of terrestrial, aerial, and satellite access into a unified network fabric. Realizing this vision demands not only advances at the Radio Access Network (RAN), but a fundamental rethinking of how the entire network infrastructure is designed, deployed, and managed. 6G envisions core networks that are intelligent, energy-efficient, and capable of adapting autonomously to heterogeneous and ever-changing demands, requirements that push beyond the capabilities of current deployments and call for entirely new orchestration mechanisms and management paradigms. Addressing these challenges requires not only advances in core network design, but also the experimental infrastructure and tools needed to develop and validate them at scale.In this thesis, we begin by establishing an empirical foundation for 5G/6G core network management through a systematic real-world evaluation of four open-source UPF implementations. By characterizing their performance and resource trade-offs under identical real 5G traffic conditions, we reveal insights that theoretical analysis cannot capture and that are essential for any subsequent management or orchestration effort.Building on this, we investigate how cloud-native principles can be applied to dynamically scale core network functions in response to real-time traffic demand. We design and evaluate a reactive scaling mechanism for Kubernetes-based 5G deployments, demonstrating that elastic core network management concretely outperforms static provisioning in both network performance and resource efficiency.We then advance toward the proactive and sustainable orchestration that 6G demands. By integrating traffic prediction with energy-aware deployment optimization across heterogeneous network function implementations, we show that intelligent orchestration can simultaneously satisfy performance requirements and reduce energy consumption, demonstrating that sustainability in the 6G core is an achievable design objective rather than an aspirational one.The fourth direction addresses one of the defining features of the 6G connectivity vision: the seamless integration of heterogeneous access technologies into a unified network fabric. 6G envisions users simultaneously
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Christakis, Sokratis
- Χριστάκης, Σωκράτης
Subjects
dc:subject × 18- Δίκτυα 5ης γενιάς
- Δίκτυα 6ης γενιάς
- Διαχείριση δικτύου πυρήνα
- Κλιμάκωση Συνάρτηση επιπέδου χρήστη
- Μεγάλα γλωσσικά μοντέλα (LLMs)
- Αλγόριθμοι μείωσης κατανάλωσης ενέργειας δικτύου
- 5G networks
- 6G networks
- Core Network Management
- Scaling User Plane Function
- Large language models
- Algorithms for network energy reduction
- 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/63117