Network Slicing Instance

The NSSAI is the fundamental identifier that allows User Equipment (UE) to request a specific slice. It consists of a collection of Single NSSAIs (S-NSSAIs) , each combining a Slice/Service Type (SST) —defining expected behavior like eMBB, URLLC, or MIoT—and an optional Slice Differentiator (SD) for further granularity. During the registration procedure, the UE provides the requested NSSAI, and the network responds with an allowed NSSAI, binding the device to authorized slice instances. This parameter set is the primary signaling mechanism for slice selection.

The Slice Orchestrator is the brain of slice lifecycle management. It automates the end-to-end instantiation, scaling, and decommissioning of a network slice instance by coordinating virtualized resources across the RAN, transport network, and 5G core. It translates high-level service requirements into concrete resource allocations, onboards Network Slice Subnet Instances (NSSIs) from sub-domain controllers, and ensures that the composite instance meets its SLA. Without an orchestrator, manual slice creation would be impossible at scale.

Slice Isolation is the non-negotiable property that makes network slicing commercially viable. It guarantees that faults, congestion, or security breaches within one slice instance cannot propagate to others sharing the same physical infrastructure. Isolation is enforced at multiple levels:

• Resource isolation: Dedicated physical resource blocks or strict scheduling quotas
• Security isolation: Separate authentication and encryption contexts
• Management isolation: Independent tenant-facing management interfaces This containment is critical for mission-critical URLLC slices coexisting with best-effort eMBB traffic.

Slice Elasticity is the dynamic ability of a network slice instance to scale its allocated virtual resources—compute, storage, and bandwidth—in response to real-time demand fluctuations. A Cloud-Native Network Function (CNF) within the slice can be horizontally scaled out during peak hours and scaled in during troughs, driven by telemetry from the Network Data Analytics Function (NWDAF) . This property ensures that the slice always meets its performance KPIs without wasteful over-provisioning, directly impacting the Slice Carbon Footprint and operational expenditure.

Slice-Aware Scheduling is a radio resource management technique at the MAC layer that differentiates treatment based on slice membership. The scheduler does not simply see a pool of users; it sees users tagged with their slice's QoS profile. It prioritizes Physical Resource Block (PRB) allocation to satisfy the Guaranteed Bit Rate (GBR) of a URLLC slice before serving a non-GBR eMBB slice. This ensures that the logical isolation defined in the core is physically enforced over the air interface, where spectrum is the scarcest resource.

A Slice SLA is the formal contract between the network operator (provider) and the vertical tenant (customer). It quantifies the performance boundaries of a network slice instance using measurable Key Performance Indicators (KPIs) :

• Throughput: Minimum guaranteed downlink/uplink data rate
• Latency: Maximum one-way delay in milliseconds
• Availability: Percentage uptime (e.g., 99.999%)
• Reliability: Packet loss rate Penalties for non-compliance are defined, making the SLA the foundational document for Slice as a Service (SlaaS) business models.