Network Analytics Tracing Entity for Unstable ID Rollback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile networks face challenges in detecting and addressing unstable analytics IDs, leading to unstable network statuses due to the lack of effective mechanisms for tracing and reverting unstable analytics outputs, which affects network performance and decision-making processes.
Innovation Solution
The introduction of a network analytics tracing entity that enables tracing and rollback of unstable analytics IDs and outputs, allowing for the identification of unstable configurations and reverting them to a stable state, ensuring stable network performance by providing rollback notifications to relevant entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the NWDAF uses black box logic to detect and fix unstable analytics IDs, then the system can maintain operational continuity, but the detection and resolution time becomes unpredictable and can range from seconds to days
Solution Approach 1:
The system performs preliminary actions by maintaining multiple versions of analytics processing logic and pre-configuring rollback mechanisms. When instability is detected, the system can immediately switch to a previous stable version without needing to collect data or wait for human intervention, thus reducing detection and resolution time while maintaining reliability
Solution Approach 2:
The system changes the parameter of analytics processing by introducing version control and rollback capability. This allows the system to transition between different versions of analytics logic based on stability requirements, enabling fast recovery from unstable states by reverting to known good versions
2Reliability
If the NF stops consuming an unstable analytics ID to prevent network deterioration, then network status can be protected, but major gaps in NF operation logic occur leading to decision-making discrepancies
Solution Approach 1:
The system segments the analytics consumption process by introducing version control and selective application of analytics logic. Different versions of analytics processing can be applied to different NFs or different time periods, allowing the system to maintain operational continuity while protecting against unstable analytics impacts through controlled segmentation of analytics application
Solution Approach 2:
The system introduces an intermediary mechanism (the rollback notification system) between the NF and the analytics processing. This intermediary allows the NF to continue operating with modified logic during unstable periods, bridging the gap between stopping analytics consumption and maintaining decision-making continuity
3Adaptability or versatility
If the NWDAF inference reselects ML models using local information or querying training, then the system can adapt to problems, but it can only become aware of problems through cycles of data collection which delays detection
Solution Approach 1:
The system implements feedback mechanisms where NFs provide information about analytics stability to the NWDAF. This feedback loop allows the NWDAF to become aware of problems faster than through data collection cycles alone, enabling timely model reselection and improving both detection speed and adaptability
Solution Approach 2:
The system performs preliminary actions by maintaining multiple ML models ready for deployment and pre-configuring the feedback collection infrastructure. When instability is reported, the system can immediately switch to alternative models without needing to collect and analyze data first, reducing detection time while maintaining adaptability
Data Source
AI summary
According to a new generation mobile network, the generation of analytics information in the mobile network, and a network analytics tracing entity and network analytics training entity configured to obtain and perform one or more rollback actions, tracing an analytics ID or one or more analytics outputs for the analytics ID, an inference or training rollback for an unstable analytics ID or for at least one unstable analytics output for the analytics ID are concerned. To this end, a network analytics tracing entity configured to: obtain an indication with information to activate a tracing of one or more analytics outputs for an analytics ID, and/or a tracing of the analytics ID, and provide a rollback notification including one or more rollback actions related to the analytics ID, if an output for the analytics ID is unstable and/or if the analytics ID is unstable.