5G L4S Traffic Scheduling Priority Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radio resource management methods in 5G cellular networks struggle to optimize the coexistence of Low Latency, Low Loss, and Scalable Throughput (L4S) traffic and non-L4S traffic, leading to increased latency for L4S packets due to inadequate visibility into PDCP and RLC queues and lack of enhanced flow control mechanisms.
Innovation Solution
Implementing enhanced radio resource management and CU-DU flow control methods that prioritize L4S traffic by adjusting scheduling priorities and frequency of data transmission, using L4S-aware schedulers and optimized timers to reduce latency and improve coexistence policies for L4S and non-L4S traffic over CU-DU mid-haul in 5G networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If current radio resource management methods are used, then non-L4S traffic is handled adequately, but L4S traffic experiences increased latency due to lack of prioritization
Solution Approach 1:
The patent segments traffic handling by implementing separate queue management for L4S and non-L4S traffic at the PDCP and RLC layers. L4S packets are routed through dedicated queues with independent scheduling parameters, allowing latency optimization for L4S without affecting non-L4S traffic handling. This segmentation enables differentiated service quality while maintaining overall system versatility.
Solution Approach 2:
The patent introduces dynamic scheduling mechanisms that adjust resource allocation based on traffic type. The MAC scheduler dynamically prioritizes L4S traffic over non-L4S traffic when L4S packets are present in the queue, while automatically adapting to maintain fair resource distribution for non-L4S traffic during different network conditions. This dynamic behavior resolves the contradiction between latency reduction and coexistence capability.
2Loss of time
If enhanced flow control is implemented for L4S traffic, then L4S latency is reduced, but system complexity increases due to additional queue management
Solution Approach 1:
The patent extends existing 5G NR queue management structures to handle both L4S and non-L4S traffic using unified data structures and scheduling frameworks. The PDCP and RLC layers are enhanced to accommodate L4S-specific parameters while maintaining backward compatibility with existing non-L4S traffic handling. This multi-functionality approach reduces complexity compared to completely separate management systems.
Solution Approach 2:
The patent introduces intermediary functions at the PDCP layer that act as a bridge between the network layer and lower layers. These intermediary functions perform L4S packet identification, marking, and queue routing without requiring fundamental changes to the underlying RLC and MAC layer structures. This intermediary approach isolates the complexity of L4S management from the core protocol stack.
3Loss of time
If L4S traffic is prioritized in scheduling, then L4S latency is reduced, but non-L4S traffic throughput may be degraded
Solution Approach 1:
The patent implements periodic scheduling opportunities for non-L4S traffic within the L4S-prioritized framework. The MAC scheduler allocates specific time slots or scheduling rounds where non-L4S traffic can access resources, ensuring periodic throughput opportunities. This periodic action prevents complete starvation of non-L4S traffic while maintaining L4S latency advantages during critical transmission periods.
Solution Approach 2:
The patent dynamically adjusts scheduling parameters such as priority weights, queue weights, and resource allocation ratios based on network conditions and traffic mix. When L4S traffic dominates, the system increases L4S priority parameters; when non-L4S traffic requires more bandwidth, the system adjusts parameters to favor throughput. This parameter adaptability resolves the contradiction between latency and throughput for different traffic types.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for optimizing coexistence of low latency, low loss and scalable throughput (L4S) traffic and non-L4S traffic in a 5G wireless system is provided, which method includes one of: a) performing one of i) a flow control between a centralized unit (CU) and a distributed unit (DU), or ii) radio resource management at the DU, to reduce latency experienced by the L4S traffic; or b) performing one of iii) a flow control between the CU and the DU to facilitate greater frequency of transmission of non-L4S traffic from CU-user plane (CU-UP) to the DU, or iv) radio resource management at the DU, to reduce a scheduling priority metric of a logical channel for the L4S traffic relative to a scheduling priority metric of a logical channel for the non-L4S traffic. To reduce latency of the L4S traffic, the L4S traffic scheduling priority is increased relative to the non-L4S traffic.