Slice Remapping

Ensures an uninterrupted user experience during the transition between slices. The process maintains the UE's IP address and ongoing Protocol Data Unit (PDU) session anchor point. This is achieved through mechanisms like make-before-break connectivity, where the new path is established before the old one is torn down, preventing packet loss and service interruption for applications like voice calls or industrial control loops.

Remapping decisions are driven by real-time energy analytics. The Network Data Analytics Function (NWDAF) predicts load changes and identifies opportunities to consolidate users onto fewer, more power-efficient slices. For example, during low traffic periods, sessions can be remapped from a standard Enhanced Mobile Broadband (eMBB) slice to a specially configured energy-saving slice that utilizes Cell Discontinuous Transmission (DTX) and Sleep Mode Coordination more aggressively.

The remapping process is governed by a strict hierarchy of policies evaluated by the Policy Control Function (PCF). These policies balance multiple objectives:

• Service Level Agreement (SLA) compliance: The target slice must meet all latency, throughput, and reliability guarantees.
• Slice Admission Control: The target slice must have sufficient resources to accept the new session without degrading existing users.
• Tenant isolation: Policies ensure a session is only moved to a slice its tenant is authorized to use.

Advanced remapping requires understanding the complex relationships between slices. A Slice Orchestrator must map dependencies to avoid cascading failures. For instance, remapping a massive IoT sensor session from a dedicated URLLC slice to a shared eMBB slice might require reconfiguring the 5G Core's User Plane Function (UPF) placement to maintain latency budgets, demonstrating the need for a holistic, topology-aware remapping strategy.

Slice remapping is a core function of a Self-Organizing Network (SON). It operates in a fully automated Observe-Orient-Decide-Act (OODA) loop:

• Observe: Monitor slice KPIs and energy consumption via telemetry.
• Orient: Analyze the data against defined policies and predictive models.
• Decide: The O-RAN Non-Real-Time RAN Intelligent Controller (Non-RT RIC) computes the optimal remapping plan.
• Act: The Near-RT RIC executes the reconfiguration via standardized E2 interface commands.

A primary goal of remapping is to minimize the network's Slice Carbon Footprint. By dynamically shifting sessions to slices hosted on infrastructure powered by greener energy sources or operating at higher capacity utilization, the overall Power Usage Effectiveness (PUE) improves. This directly links a real-time network control action to a quantifiable sustainability metric, enabling operators to meet environmental, social, and governance (ESG) targets.

A MAC-layer mechanism that prioritizes radio resource allocation based on slice membership. When a UE is remapped to a new slice, the scheduler immediately adjusts its scheduling weight, priority queue, and resource block allocation.

• Ensures a remapped URLLC session receives immediate priority over eMBB traffic
• Dynamically adjusts Modulation and Coding Scheme (MCS) selection per slice policy
• Coordinates with Adaptive Bandwidth Part (BWP) switching for power-efficient remapping
• Prevents resource starvation during mass remapping events through weighted fair queuing

The standardized 5G parameter set that enables UE and network to negotiate slice selection. During remapping, the Configured NSSAI or Allowed NSSAI is updated to reflect the new target slice.

• S-NSSAI uniquely identifies each slice instance via Slice/Service Type (SST) and Slice Differentiator (SD)
• The AMF validates NSSAI changes against subscription data during session modification
• Supports default subscribed S-NSSAI fallback when remapping to a best-effort slice
• Enables energy-aware slice selection by incorporating power efficiency metrics into the selection policy

The automation framework that triggers and executes slice remapping without human intervention. A policy-driven controller continuously monitors KPIs, detects optimization opportunities, and orchestrates the reassignment.

• Monitor: NWDAF collects real-time slice load, latency, and energy telemetry
• Analyze: AI/ML models predict SLA violations or energy inefficiency windows
• Decide: Policy engine determines optimal target slice and remapping timing
• Execute: Slice Orchestrator coordinates session transfer across RAN, transport, and core domains
• This loop enables sub-second remapping decisions for latency-critical URLLC sessions

A policy-driven function that steers UE sessions to the most energy-efficient slice instance capable of satisfying service requirements. This is the primary trigger for energy-motivated slice remapping.

• Evaluates the Slice-Level Energy Model to compare power consumption across candidate slices
• Factors in Power Usage Effectiveness (PUE) of the data centers hosting slice workloads
• Coordinates with Sleep Mode Coordination to consolidate sessions onto fewer active slices
• May temporarily remap non-GBR sessions to a lower-power slice during off-peak hours
• Balances energy savings against Slice SLA compliance and user experience metrics

The ability of a network slice to dynamically scale its virtualized resources in response to remapping events. When sessions are remapped into a slice, it must scale up; when sessions leave, it should scale down to conserve energy.

• Horizontal scaling: Spinning up additional CNF instances to handle increased load
• Vertical scaling: Allocating more vCPU or memory to existing network functions
• Dynamic Voltage and Frequency Scaling (DVFS) adjusts compute power in real-time
• Resource Overbooking policies may relax during remapping surges to maintain QoS
• Elasticity ensures the target slice doesn't become a bottleneck after mass remapping