Back to search

University of Thessaly (UTH)

Network reconfiguration and management in 6g telecommunication networks

Abstract

dc:description

Fifth-generation (5G) telecommunication networks promise unprecedented improvements in connectivity, offering ultra-high data rates, ultra-reliable low-latency communication (URLLC), and massive connectivity of diverse devices. These capabilities are essential for enabling transformative applications such as autonomous vehicles, remote surgery, industrial automation, and the Internet of Things (IoT). However, despite the theoretical advances, practical deployments frequently fail to meet anticipated performance benchmarks. This performance gap primarily arises from simplistic assumptions regarding user mobility patterns, static resource allocation strategies, and limited adaptability to changing network conditions. As the industry transitions toward sixth-generation (6G) networks, addressing these challenges through dynamic reconfiguration and advanced management mechanisms becomes critically important. This thesis investigates fundamental questions related to dynamic resource allocation, seamless service continuity during user mobility, and robust security in highly automated edge network environments. Specifically, the research addresses the following key problems: how can Artificial Intelligence and Machine Learning (AI/ML) techniques be leveraged to dynamically enhance and optimize resource allocation within the Radio Access Network (RAN)? How can low-latency access be continuously maintained for mobile users within Multi-Access Edge Computing (MEC) frameworks? Lastly, how can the security and resilience of edge network infrastructures be guaranteed against evolving threats such as Denial of Service (DoS) attacks? To tackle dynamic resource allocation challenges, we designed and incorporated an MLOps platform within the 5G architecture, which collects user and radio data through a custom Network Data Analytics Function (NWDAF) module. By employing deep learning techniques, we predict user demands, application interactions, and radio conditions, enabling proactive reallocation of Physical Resource Blocks (PRB). This approach demonstrates a successful transition from traditional network-aware applications—where services must adapt to network conditions—to a novel service-aware paradigm, wherein the network autonomously aligns itself with real-time application needs, significantly reducing resource over- and under-provisioning. To ensure seamless service continuity in MEC environments, we introduced a heterogeneous connectivity framework integrating both 3GPP and non-3GPP technologies. We developed a service migration controller that dynamically selects optimal Radio Access Technology (RAT) paths based on real-time radio conditions, that also migrates edge services closer to user locations. Additionally, we proposed and implemented a Deep Reinforcement Learning (DRL)-based migration approach, utilizing multi-cell Round-Trip Time (RTT) measurements to proactively reposition services and maintain continuous, low-latency access during user mobility. Extensive experimental evaluations validated the effectiveness of this approach, demonstrating uninterrupted high-quality user experiences.Addressing network security, we integrated a robust anomaly detection mechanism within an Open RAN architecture. This system identifies and mitigates real-time security threats such as DoS attacks, dynamically adjusting resource allocations, and manages the users to maintain network integrity and service quality. The synergy of dynamic resource allocation and security enhancements significantly improves the resilience and reliability of next-generation network infrastructures. Comprehensive experimental evaluations conducted in realistic testbed environments highlight substantial improvements across multiple performance metrics, including reduced latency, increased throughput, optimized resource utilization, and enhanced energy efficiency. These results demonstrate the practical viability and efficacy of the proposed methodologies, providing robust foundations for addressing similar challenges anticipated in emerging 6G network ecosystems.

Degree

thesis:*
Grantor dc:publisher
University of Thessaly (UTH)
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Tsourdinis, Theodoros
  • Τσουρδίνης, Θεόδωρος

Subjects

dc:subject × 17

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/60099

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Tsourdinis, Theodoros; Τσουρδίνης, Θεόδωρος. Network reconfiguration and management in 6g telecommunication networks. University of Thessaly (UTH), 2025. http://hdl.handle.net/10442/hedi/60099