PSACCF: Prioritized Online Slice Admission Control Considering Fairness in 5G/B5G Networks

5G/B5G is intended to support various services with the assistance of network slices, and each slice requires adequate resources to provide the prenegotiated service quality. The Slice Admission Control (SAC) algorithm is a necessity for Slice Providers (SPs) to guarantee the Quality of Service (QoS) and Quality of Experience (QoE) of each admitted request. In such circumstances, concerns regarding the priority of services and the fairness of resource allocation are meaningful topics. However, these two issues do not receive sufficient research attention simultaneously in the literature. Here, we study the SAC problem in 5G/B5G networks, aiming to enhance the fairness degree while satisfying the necessary priority requirements. We first reinterpret priority as a higher Cumulative Service Acceptance Ratio (CSAR) and adopt the uniformity of adjacent CSAR gaps to reflect fairness. The SAC problem is formulated as a nonlinear and nonconvex multiobjective optimization problem. Thus, we propose a heuristic algorithm called Prioritized Slice Admission Control Considering Fairness (PSACCF) to solve it. This approach introduces the resource efficiency of services to amend priority violations and then promotes fairness by setting the target CSARs for each service type and improving their actual CSARs. Numerous simulations are conducted to compare the performance of PSACCF with that of Moderate High Priority First (MHPF) and Aggressive High Priority First (AHPF). The results show that PSACCF can achieve at least a 33.6% improvement in fairness degree and a higher minimum average resource utilization while maintaining a nearly identical priority indicator to that of the compared methods.

I. Introduction and Motivations

Future applications of 5G are believed to belong to three broad categories [1] namely, enhanced Mobile Broadband (eMBB), ultra-Reliable Low Latency Communication (uRLLC), and massive Machine Type Communication (mMTC). These categories focus on different performance indicators, and the types and amounts of resources consumed are also distinct. Specifically, eMBB requires support for high mobility and large communication bandwidth, uRLLC has strict latency and reliability limits, and mMTC requires adequate access capacity. To handle the heterogeneity of these applications, the network slice [2], [3] is employed by 5G/B5G networks to perform customized services, resulting in a new paradigm that enables the deployment of virtual networks on generic equipment. By designing and adjusting the structure of slices, slice providers can establish logic networks that possess the corresponding abilities required by Slice Tenants (STs).