Joint User-Slice Pairing and Association Framework Based on H-NOMA in RAN Slicing

Area of Science:

• * Wireless communication networks
• * Network architecture and management
• * Resource allocation and optimization

Background:

• * Radio Access Network (RAN) slicing is crucial for flexibility and isolation in multiservice cellular networks.
• * Existing RAN slicing approaches inadequately address heterogeneous network (HetNet) architectures.
• * Increasing mobile traffic demands diverse Quality of Service (QoS) levels, including ultra-reliability low-latency communications (uRLLC) and enhanced mobile broadband (eMBB).

Purpose of the Study:

• * To propose a pairing-network slicing (NS) approach for Multiservice RAN within a 5G HetNet architecture.
• * To optimize resource allocation considering QoS, baseband resources, fronthaul/backhaul capacities, and slice isolation.
• * To develop a framework for joint user-slice pairing and association to maximize data rates and meet stringent latency/reliability requirements.

Main Methods:

• * Development of a RAN slicing framework for joint user equipment (UE) association.
• * Implementation of multicell non-orthogonal multiple access (NOMA) with successive interference cancelation (SIC) for resource allocation.
• * Formulation of UE-slice pairing and association as optimization problems, utilizing a one-to-many matching game (UE-slice association algorithm) and a stable pairing algorithm (UE-slice pairing algorithm).

Main Results:

• * The proposed approach effectively addresses RAN slicing criteria, ensuring inter- and intra-slice isolation.
• * Joint user-slice pairing and association maximize enhanced mobile broadband (eMBB) UE data rates while satisfying ultra-reliability low-latency communications (uRLLC) requirements.
• * Numerical analysis confirms the optimization of system throughput and reduction in uRLLC latency through mini-slot association and pairing.

Conclusions:

• * The proposed pairing-network slicing approach offers a viable solution for multiservice RAN in 5G HetNets.
• * The developed algorithms ensure stable UE-slice connections and efficient spectrum coexistence for different services.
• * This framework successfully balances competing QoS demands, enhancing overall network performance and user experience.