QoS-Aware Balancing

A core mechanism is the differential treatment of Guaranteed Bit Rate (GBR) and Non-GBR bearers. When predictive models forecast cell overload, the balancing algorithm preemptively shifts Non-GBR traffic to adjacent cells to preserve the Guaranteed Flow Bit Rate (GFBR) for GBR flows on the source cell. This is not reactive congestion control; it is a proactive reservation of Physical Resource Blocks (PRBs) based on forecasted demand from latency-intolerant services.

In a 5G network slicing architecture, each slice is a logical network with its own SLA. QoS-aware balancing operates at the slice level, ensuring that the load state of one slice does not degrade another. Key metrics include:

• Slice-level PRB utilization: Prevents a single slice from starving others of radio resources.
• Slice-specific prediction models: A URLLC slice model forecasts short-term, bursty demand, while an eMBB slice model forecasts longer-term, high-throughput patterns.
• Cross-slice policy enforcement: Ensures that traffic steering for an eMBB slice never violates the latency SLA of a co-located URLLC slice.

The ultimate objective is not just meeting network KPIs but ensuring Quality of Experience (QoE) . The steering engine translates QoE metrics into actionable thresholds:

• Video streaming: Predicts stall probability based on forecasted throughput and buffer status. If the predicted throughput drops below the encoding bitrate, the flow is proactively shifted.
• VoNR/VoLTE: Monitors the predicted Mean Opinion Score (MOS). If the MOS is forecasted to drop below 3.5, the call is handed over to a cell with a stronger CQI profile.
• Web browsing: Uses predicted round-trip time (RTT) to ensure interactive responsiveness.

A scalar value that defines a specific QoS characteristic for a 5G flow, including resource type (GBR, Non-GBR, Delay-Critical GBR), priority level, packet delay budget, and packet error rate. The 5QI is the primary label that QoS-aware balancing algorithms inspect to classify and prioritize traffic. Standardized 5QI values map to well-known services—for example, 5QI 3 for real-time gaming (50ms delay budget) and 5QI 9 for TCP-based web browsing (300ms delay budget).

A fundamental QoS distinction that determines how a QoS-aware balancer allocates resources:

• GBR flows: Require a fixed minimum bit rate, even under congestion. Used for voice calls, live video, and V2X communications. The network must reserve resources before admitting the flow.
• Non-GBR flows: Best-effort delivery with no minimum rate guarantee. Used for web browsing, email, and file downloads. These flows are the first to be throttled or shifted during load balancing.

A QoS-aware balancer must satisfy all GBR commitments before distributing remaining resources among Non-GBR flows.

An upper bound on the time a packet may be delayed between the User Equipment and the User Plane Function. The PDB is a critical constraint for QoS-aware balancing:

• Ultra-Reliable Low-Latency Communication (URLLC) flows may have a PDB as low as 0.5ms
• Enhanced Mobile Broadband (eMBB) flows typically tolerate 10–100ms

When a predictive balancer forecasts congestion on a cell, it must proactively offload delay-sensitive flows to cells where the end-to-end latency budget can still be met, accounting for handover interruption time and backhaul latency.

The process of associating IP flows with specific QoS flows and mapping them to Data Radio Bearers (DRBs). A QoS-aware balancer must understand this mapping hierarchy:

• QoS Flow ID (QFI): Identifies a specific QoS flow within a PDU Session
• Service Data Adaptation Protocol (SDAP): Maps QoS flows to DRBs
• Reflective QoS: Allows the UE to derive uplink QoS rules from downlink traffic patterns

When steering traffic, the balancer must ensure that all flows belonging to the same PDU Session maintain session continuity and that QoS flow-to-DRB remapping occurs seamlessly at the target cell.

An extension of QoS-aware balancing that considers Network Slice membership alongside per-flow QoS requirements. Each network slice (e.g., eMBB slice, URLLC slice, mMTC slice) has a distinct Service Level Agreement (SLA) with specific performance targets.

A slice-aware balancer must:

• Ensure that load shifting does not violate the SLA of any slice
• Prioritize intra-slice load redistribution before inter-slice resource borrowing
• Respect slice isolation policies, especially for dedicated or prioritized slices

This requires the balancer to ingest Network Slice Selection Assistance Information (NSSAI) and correlate it with real-time slice-level KPI monitoring.

The integration of QoS constraints into the handover decision process. Traditional handover algorithms use only signal strength (RSRP/RSRQ), but QoS-aware handover adds service-level criteria:

• QoS-based A5 event: Trigger handover when serving cell quality drops below a threshold AND neighbor cell can satisfy the flow's PDB
• Admission control gating: Prevent handover to a target cell that lacks sufficient resources to maintain the flow's GBR
• QoS Class Identifier (QCI) prioritization: In LTE, prioritize handover for QCI 1 (conversational voice) over QCI 9 (background data)

This prevents the scenario where a UE successfully hands over but immediately experiences service degradation due to resource starvation.