RIC-Based Stale UE Context Detection in O-RAN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In 5G networks, there is a need for an improved system and method to manage resource allocation mismatches, particularly for UE contexts, which can lead to stale contexts and resource exhaustion, causing critical service outages.
Innovation Solution
The proposed solution utilizes an RIC-based technique to detect and clean up stale RAN resources. This involves periodic polling of RAN nodes, telemetry data analysis, performance measurement counters, and resource status reporting to identify and correct mismatches in UE contexts across different RAN nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If resource allocation is managed across distributed RAN nodes without centralized coordination, then network autonomy and distributed operation are improved, but resource allocation mismatches and stale UE contexts occur leading to service outages
Solution Approach 1:
The patent implements a feedback mechanism where the RIC receives resource allocation information from multiple RAN nodes, analyzes mismatches, and sends correction commands back to the nodes. This closed-loop feedback system enables autonomous detection and resolution of stale UE contexts while maintaining distributed operation, thus resolving the contradiction between distributed autonomy and service reliability.
Solution Approach 2:
The RIC acts as an intermediary between distributed RAN nodes, coordinating resource allocation and detecting mismatches. This intermediary layer enables centralized coordination without direct node-to-node communication, allowing distributed nodes to operate autonomously while preventing resource allocation conflicts that would compromise service continuity.
2Measurement precision
If periodic polling and telemetry analysis are implemented to detect stale UE contexts, then detection accuracy is improved, but signaling overhead and network load increase
Solution Approach 1:
The system performs preliminary actions by configuring RAN nodes to report resource allocation information at specific events (UE context release, handover completion) before stale contexts actually occur. This event-driven reporting reduces the need for continuous periodic polling, maintaining high detection accuracy while minimizing signaling overhead and energy consumption.
Solution Approach 2:
The patent implements periodic polling at optimized intervals rather than continuous monitoring. The RIC periodically requests resource allocation status from RAN nodes at configurable time intervals, balancing detection accuracy with signaling overhead. This periodic action reduces network load compared to continuous monitoring while still effectively identifying stale UE contexts.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for detecting stale UEs in O-RAN systems based on combining information derivable from the Radio Intelligent Controller consisting in a radio intelligent controller (RIC) polling a centralized unit control plane (CU-CP), a centralized unit user plane (CU-UP), and a distributed unit (DU) to receive information regarding allocated RAN resources associated with a user equipment (UE); and correlating, by the RIC, the received information regarding the allocated RAN resources associated with the UE to determine whether the CU-CP, CU-UP and DU are in sync with respect to the allocated RAN resources associated with the UE. A discrepancy in the allocated RAN resources among the CU-CP, CU-UP and DU indicates stale RAN resource allocation by at least one of the CU-CP, CU-UP and DU.