NRF Service Discovery Preference for Combined and Standalone SMFs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 5G network architectures lack efficient methods to prioritize combined PGW-C+SMF or standalone SMF during PDU session establishment, leading to suboptimal discovery and increased processing time due to mixed responses and internal filtering requirements.
Innovation Solution
Introduce a new information element, 'preferred-pgw-ind', to indicate preference for combined PGW-C+SMF or standalone SMF, allowing the Network Repository Function (NRF) to prioritize and respond with preferred network functions, reducing the need for additional filtering and ensuring optimal node selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the NRF returns mixed responses of combined and standalone SMFs without preference indication, then the discovery process is simple, but the service consumer requires additional internal filtering and sorting based on priorities, increasing processing time
Solution Approach 1:
The NRF performs preliminary sorting and prioritization of SMF candidates before returning them to the service consumer. By pre-organizing the candidate list according to priorities, capacity, load, and other criteria, the NRF eliminates the need for the service consumer to perform additional filtering and sorting operations, thereby reducing processing time and improving overall PDU session establishment efficiency
2Measurement precision
If the NRF performs internal sorting and selection of the most matched SMF, then the service consumer gets optimal results, but the NRF processing complexity and time increase
Solution Approach 1:
The invention introduces a new information element 'preferred-pgw-ind' that changes the parameter space of the discovery request. This parameter enables the service consumer to explicitly indicate preference for combined PGW-C+SMF or standalone SMF, allowing the NRF to filter results based on this preference without requiring complex multi-criteria optimization algorithms, thus maintaining matching accuracy while reducing processing complexity
3Adaptability or versatility
If the NRF returns all discovered SMF candidates without preference filtering, then the service consumer has more options, but the response size and processing overhead increase
Solution Approach 1:
The NRF extracts and returns only the most relevant SMF candidates that match both the query parameters and the indicated preference. By applying the 'preferred-pgw-ind' filter, the NRF extracts the subset of SMFs that are both compatible and preferred, reducing the response size while maintaining adaptability for the service consumer to select from qualified candidates
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The embodiments herein relate to support of preferred service provider/producer query for Network Repository Function (NRF). In some embodiments, there proposes a method (400) performed by a first network function (101, 201) implementing a service consumer. In an embodiment, the method (400) may comprise the step of transmitting (S401), to a second network function (102) implementing a NRF, adiscovery request for discovering one or more third network functions (103) implementing service producers. The discovery request may include a preference indication for indicating whether one or more combined third network functions (103) or one or more standalone third network functions (103) are preferred. The method (400) may further comprise the step of receiving (S402), from the second network function (102), adiscovery response including information about the discovered one or more third network functions (103). The discovered one or more third network functions (103) may be provided based on the preference indication by the second network function (102). With the embodiments herein, the combined PGW-C+SMF and/or standalone SMF with priority may be presented in the query results for service provider/producer, and may reduce the processing time ofboth service consumer and the NRF.