UPF Selection Using NWDAF Analytics for Edge Path Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The selected user plane path between the UE and the edge computing server is not an optimal solution, leading to suboptimal user experience and service performance in 5G-based distributed cloud infrastructure.
Innovation Solution
The SMF determines the UPF by analyzing user experience and performance data from the NWDAF, considering factors like UE location, application position, and user plane anchor positions to select an optimal UPF for edge computing services, optimizing the user plane path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the SMF selects a UPF based on basic parameters like UE position and network topology, then the path establishment is simple and fast, but the user experience and service performance are suboptimal
Solution Approach 1:
The system performs preliminary actions by having the NWDAF pre-analyze network data and generate user experience analytics and performance analytics before the SMF needs to select a UPF. These analytics results are stored and can be quickly retrieved when needed, avoiding complex real-time analysis while still achieving optimal path selection.
Solution Approach 2:
The NWDAF acts as an intermediary between the SMF and the raw network data. It collects, processes, and analyzes network performance data, then provides processed analytics results to the SMF. This intermediary handles the complexity of data analysis, allowing the SMF to make informed decisions without directly implementing complex analysis algorithms.
2Reliability
If the SMF collects and analyzes comprehensive network data in real-time to select the optimal UPF, then the user experience is optimized, but the signaling overhead and processing time increase
Solution Approach 1:
The NWDAF performs data collection and analysis in advance, generating analytics results that are stored for future use. When the SMF needs to select a UPF, it can directly query these pre-computed results rather than collecting and analyzing raw data in real-time, significantly reducing selection time while maintaining optimization quality.
Solution Approach 2:
Instead of the SMF directly accessing and analyzing raw network data, it uses copies of processed analytics results generated by the NWDAF. These analytics results contain the essential information needed for optimal UPF selection without requiring the SMF to handle the full complexity of raw data analysis.
3Productivity
If the system uses traditional DNS-based application position determination, then the implementation is simple, but it cannot provide optimal user plane path selection for edge computing services
Solution Approach 1:
The system changes the parameters used for application position determination from simple DNS-based geographic location to comprehensive analytics including user experience metrics, network performance data, and application-specific requirements. This enables optimal UPF selection that considers multiple factors beyond basic location matching.
Solution Approach 2:
The NWDAF serves as an intermediary that provides enhanced application position and performance information to the SMF, going beyond what traditional DNS can provide. It analyzes network data to determine the optimal application server position and characteristics, enabling more informed UPF selection decisions.
Data Source
AI summary
The present disclosure provides UPF determination method and device, information providing method and device, and media. An NWDAF receives a data analytic request about an application from an SMF and/or an AF, analyzes application run data collected from the SMF and the AF to generate a user experience analytic result and/or a performance analytic result including a user experience analysis and/or a performance analysis for the application corresponding to a location area, the location area including at least one of a UE location, an area where the application locates, or an area where a user plane anchor locates, and provides a data analytic result to the SMF and/or the AF, so that the SMF optimizes a user plane path in accordance with the analytic result and the AF adjusts a position of a target position in accordance with the analytic result.