Quality Of Service Aware User Association For Efficient Network Slicing In Multi-Tenant 5g Networks

Swathi, Subbiah Jayanthi (2022) Quality Of Service Aware User Association For Efficient Network Slicing In Multi-Tenant 5g Networks. Master dissertation/thesis, UTAR.

Preview

PDF Download (1796Kb) | Preview

Abstract

Network slicing is one of the key technologies for fifth generation (5G) communications to overcome hurdles such as increased network density and data traffic due to the growing number of mobile devices. Multi-tenant network slicing allows multiple Mobile Virtual Network Operators (MVNOs) to share the physical network infrastructure, each known as a tenant, for service provisioning. In particular, multi-tenant network slicing for heterogeneous networks (Het-Nets), where small cells co-exist within macro cells, has received significant attention due to the advantages of Het-Nets in boosting network capacity and coverage. Despite a great amount of research in multi-tenant network slicing for Het-Nets, user association for multi-tenant network slicing in Het-Nets has not well investigated. A few studies have delved into user association for multi-tenant network slicing in Het-Nets, but various performance aspects, namely fairness, quality of service (QoS), user admission, power consumption are not jointly considered. The main objective of this thesis is to study and develop a user association scheme that considers the aforementioned performance aspects for multi-tenant network slicing. iii In this thesis, the user association problem for network slicing in multi-tenant 5G Het-Nets is investigated under network slicing scenarios with single and multiple base stations (BSs) per MVNO. In the first part of the thesis, a constrained user association problem that maximizes the weighted sum throughput is formulated under the network slicing scenario with single BS per MVNO. To solve the problem, it is converted into an unconstrained problem and is solved using Genetic Algorithm (GA). Overall, in terms of fairness, QoS, percentage of users associated and the power consumption, the proposed scheme outperforms the existing scheme. In the second part of this thesis, a user association problem that maximizes the weighted sum throughput is formulated under the network slicing scenario with multiple BSs per MVNO, which has further taken into account intra-slice interference. The problem is again solved using the penalty function approach and GA. The proposed GA based user association scheme outperforms the existing Signal to Interference plus Noise Ratio (SINR) based user association scheme with increased fairness, QoS, number of users associated to the network, and decreased power consumption.

Actions (login required)