5G Core Network Function Data Path Discovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing 5G Core network's data path discovery for user plane functions (UPFs) is suboptimal due to the failure to consider increased load on previously present UPFs and latency between identified UPFs, leading to non-optimal data paths.

Innovation Solution

The proposed solution involves the Network Repository Function (NRF) obtaining an overview of available optimal paths by registering and periodically updating the load information of each UPF. The Session Management Function (SMF) queries the NRF for the complete optimal path between the Radio Access Network (RAN) and the Packet Data Unit Session Anchor (PSA), considering network topology, load, and relative capacities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If individual UPF discovery and insertion is performed sequentially by SMF querying NRF, then UPF selection is made based on local information, but the data path becomes non-optimal due to failure to consider overall network topology and load distribution

Engineering Contradiction:
ImproveUPF discovery and insertion processVSAvoiddata path optimality
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The NRF is enhanced to act as an intermediary that maintains global knowledge of network topology and UPF load information. When SMF queries NRF for UPF discovery, the NRF uses this global view to compute and return the optimal complete data path rather than just individual UPF selections, thereby resolving the contradiction between operational simplicity and path optimality.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The solution transitions from sequential single-UPF selection to holistic multi-UPF path optimization by adding the dimension of global network topology awareness. The NRF computes paths considering multiple UPFs simultaneously and their interconnections, moving beyond the limited local decision-making of individual SMF queries.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If multiple UPFs are inserted sequentially in the data path as UE moves between regions, then service continuity is maintained, but the number of UPFs increases leading to increased GTP overhead and reduced efficiency

Engineering Contradiction:
Improveservice continuity during UE movementVSAvoidGTP overhead
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary computation of the complete optimal data path at the NRF before UE movement issues arise. By pre-calculating the entire path considering all factors including load and topology, the system avoids reactive sequential UPF insertions that would increase overhead, while still maintaining service continuity through the pre-planned path.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The NRF continuously maintains updated information about network topology and UPF load conditions. This feedback mechanism allows the NRF to compute optimal paths that minimize the number of UPFs required, thereby reducing GTP overhead while ensuring service continuity is maintained through the optimized path.

Inventive Principle:
Principle #23Feedback

3Device complexity

If UPFs are selected without considering their present load, then UPF selection is simplified, but load imbalance occurs with increased load on previously present UPFs

Engineering Contradiction:
ImproveUPF selection processVSAvoidload distribution efficiency
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The solution incorporates UPF load information as a dynamic parameter in the path computation process. The NRF queries UPF load conditions and uses this parameter alongside topology information to compute optimal paths that distribute load efficiently, thereby resolving the contradiction between selection simplicity and load distribution efficiency.

Inventive Principle:
Principle #35Parameter changes

4Productivity

If the complete optimal path is computed by NRF considering overall network topology and load, then data path optimality is improved, but the complexity of path computation and information management increases

Engineering Contradiction:
Improvedata path optimalityVSAvoidNRF computation and information management
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The NRF is enhanced to perform multiple functions: maintaining global topology information, tracking UPF load conditions, computing optimal paths, and providing this information to SMF. By consolidating these functions in a single network entity, the solution achieves path optimality without distributing the computational complexity across multiple nodes.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3909284B1Optimal network function data path discovery for 5g core
Publication Date: 2025.04.02 MAVENIR NETWORKS INC
  • EP3909284B1 patent drawingFigure 1
  • EP3909284B1 patent drawingFigure 2
  • EP3909284B1 patent drawingFigure 3

AI summary

In a method and an apparatus for optimizing network function (NF) data path discovery for 5G Core network operation, each NF registers itself with the NRF and periodically updates the NRF about the NF's present load, thereby giving the NRF complete information regarding the network topology, the NFs' present load, and the relative capacities of the NFs, which information items can be used in determining the complete optimal path from the RAN to the PSA. When an SMF wants to insert an Intermediate UPF (l-UPF) into a data path, instead of querying the NRF for a UPF serving the present geographical area (e.g., of the User Equipment (UE)) and deciding on the UPF insertion locally based on the present geographical area, the SMF queries the NRF for the complete optimal path. The l-UPF load can be taken into consideration in calculating the optimal path between the RAN and the PSA.