Value-Added Functions in 5G Cellular Systems and Associated RAN Resource Sharing
Yarkina, Natalia (2025)
Tampere University
2025
Tieto- ja sähkötekniikan tohtoriohjelma - Doctoral Programme in Computing and Electrical Engineering
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Väitöspäivä
2025-08-29
Julkaisun pysyvä osoite on
https://urn.fi/URN:ISBN:978-952-03-4008-7
https://urn.fi/URN:ISBN:978-952-03-4008-7
Tiivistelmä
The specifications of the fifth-generation cellular networks (5G) have laid the basis for the future evolution of cellular systems towards a ubiquitous all-purpose communication infrastructure. Along with important quantitative advances in terms of data rates, latency performance and coverage, the standard dictates a paradigm shift towards great flexibility in the supported services and business models, network topology and control. This flexibility is enabled by several key value-added 5G functions, among which Network Slicing, Integrated Access and Backhaul (IAB), and broadcast/multicast communications.
Network Slicing ensures the versatility of 5G networks and their capability to support highly heterogeneous use cases and traffic types. Multicast communications are intended for resource-efficient delivery of video content, such as television broadcast communications. IAB enables cost-efficient network densification necessary for the wide deployment of high-band spectrum. Although the purposes of these technologies are different, they all impact the operation of the radio access network (RAN), the most complex and resource-constrained component of the 5G infrastructure, posing novel specific research challenges.
This dissertation addresses the aforementioned value-added 5G features from the perspective of RAN resource sharing. Firstly, it proposes a flexible prioritybased scheme for inter-slice resource arbitration in a virtualized 5G RAN, which provides slice performance isolation, fairness and efficient resource utilization. Depending on the parameter settings, the scheme reduces the session drop probability by an order of magnitude and improves the average user satisfaction by up to 90 % compared to static slicing. A session-level continuous-time Markov model is employed for the scheme’s performance evaluation. Also, to accelerate resource arbitration, a machine-learning-based enhancement is proposed.
Secondly, to efficiently share the time resource among conflicting links of a halfduplex multi-hop IAB network, two low-complexity delay-oriented link scheduling policies are devised. Simulation results indicate that the proposed policies deliver stably low packet delays in low-to-medium traffic conditions covering above 60 % of the system’s capacity region. To cover the capacity region in full, a hybrid control method combining fixed delay-oriented link scheduling with throughput-optimal control is also proposed.
Finally, the resource efficiency of multicast shared delivery in IAB systems with highly directional transmissions is investigated. Multicasting is shown to save up to 40 % of IAB RAN resources and can therefore be recommended along with multicast- and backhaul-aware user association. The dissertation employs the mathematical tools of the queueing theory, convex and combinatorial optimization, and dynamic programming.
Network Slicing ensures the versatility of 5G networks and their capability to support highly heterogeneous use cases and traffic types. Multicast communications are intended for resource-efficient delivery of video content, such as television broadcast communications. IAB enables cost-efficient network densification necessary for the wide deployment of high-band spectrum. Although the purposes of these technologies are different, they all impact the operation of the radio access network (RAN), the most complex and resource-constrained component of the 5G infrastructure, posing novel specific research challenges.
This dissertation addresses the aforementioned value-added 5G features from the perspective of RAN resource sharing. Firstly, it proposes a flexible prioritybased scheme for inter-slice resource arbitration in a virtualized 5G RAN, which provides slice performance isolation, fairness and efficient resource utilization. Depending on the parameter settings, the scheme reduces the session drop probability by an order of magnitude and improves the average user satisfaction by up to 90 % compared to static slicing. A session-level continuous-time Markov model is employed for the scheme’s performance evaluation. Also, to accelerate resource arbitration, a machine-learning-based enhancement is proposed.
Secondly, to efficiently share the time resource among conflicting links of a halfduplex multi-hop IAB network, two low-complexity delay-oriented link scheduling policies are devised. Simulation results indicate that the proposed policies deliver stably low packet delays in low-to-medium traffic conditions covering above 60 % of the system’s capacity region. To cover the capacity region in full, a hybrid control method combining fixed delay-oriented link scheduling with throughput-optimal control is also proposed.
Finally, the resource efficiency of multicast shared delivery in IAB systems with highly directional transmissions is investigated. Multicasting is shown to save up to 40 % of IAB RAN resources and can therefore be recommended along with multicast- and backhaul-aware user association. The dissertation employs the mathematical tools of the queueing theory, convex and combinatorial optimization, and dynamic programming.
Kokoelmat
- Väitöskirjat [5325]