CU-DU Overload Status Signaling for Faster Radio Scheduling Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cellular networks experience overload, leading to poor performance and inefficient resource usage, with existing overload control strategies causing packet discarding and elevated operating costs due to slow signaling and resource wastage.
Innovation Solution
Implementing a system that determines an overload state of a centralized unit user plane and transmits overload status data to a distributed unit, using new radio user plane or F1 application protocol data to facilitate deprioritization and scheduling adjustments, minimizing resource waste and improving network efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If overload status data is transmitted to distributed unit for faster mitigation, then network response speed improves, but signaling complexity increases
Solution Approach 1:
The system divides the gNB into centralized unit (CU) and distributed unit (DU) components. The CU handles high-level control while the DU handles radio frequency operations and local overload mitigation. This segmentation allows the DU to autonomously respond to overload conditions without requiring complex end-to-end signaling, thus improving response speed while managing signaling complexity through functional decomposition.
Solution Approach 2:
The system pre-configures the DU with overload mitigation capabilities and authority. When overload status data is received from the CU, the DU can immediately execute pre-planned mitigation actions (such as rejecting new connections or prioritizing existing ones) without waiting for further signaling or approval. This preliminary preparation enables faster response while keeping signaling simple.
2Reliability
If packets are discarded during overload to protect network, then network stability improves, but resource utilization deteriorates
Solution Approach 1:
Instead of discarding all packets during overload, the system applies partial action by selectively managing different packet types or flows. The DU can prioritize critical traffic while rejecting non-essential connections, or apply gradual throttling rather than complete rejection. This partial action maintains network stability for essential services while reducing resource waste compared to blanket packet discarding.
Solution Approach 2:
The system applies different quality treatments to different packets or connections based on local conditions at the DU. Rather than uniform packet discarding, the DU can identify and protect high-priority traffic while allowing lower-priority traffic to be rejected. This local quality differentiation maintains stability for important communications while minimizing overall resource waste.
3Measurement precision
If centralized unit controls overload mitigation, then control precision improves, but response time deteriorates
Solution Approach 1:
The system segments the overload control function into two parts: the CU performs precise overload detection and status determination using aggregated network-wide information, while the DU performs rapid local mitigation execution. This segmentation allows the CU to maintain precise control decisions while the DU provides fast response, eliminating the need for the CU to directly execute each mitigation action and thus reducing response time.
Solution Approach 2:
The system uses a simple status indication message as an intermediary between the CU and DU. The CU determines overload status with precision and sends a compact status indicator to the DU, which then autonomously executes mitigation. This intermediary mechanism transfers control information efficiently without requiring complex real-time signaling, maintaining both detection precision and response speed.
Data Source
AI summary
Overload status data transmission to a distributed unit, enabling overload action at distributed unit (e.g., using a computerized tool), is enabled. For example, a system can comprise: a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising: determining an overload state of a centralized unit user plane of network equipment, and transmitting overload status data, representative of the overload state of the centralized unit user plane, to a distributed unit of the network equipment.