AI/ML Maintenance Orchestration for Mobile Core Network Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for maintaining cloud-based mobile core network functions require manual intervention from multiple teams, leading to inefficiencies, errors, and potential disruptions due to complex dependencies between network functions, especially in 5G networks, which are not addressed by existing 3GPP standards.
Innovation Solution
A machine learning-assisted maintenance orchestrator uses AI/ML techniques to identify and manage dependencies between network functions, enabling semi-autonomous deactivation and reactivation with minimal disruption, by discovering shared cloud infrastructure failure domains and defining logical sequences for network function operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual intervention from multiple teams is used for maintenance, then network function dependencies can be managed, but maintenance efficiency is low and errors occur frequently
Solution Approach 1:
The system implements self-service through automated dependency discovery and maintenance orchestration. The AI/ML model automatically identifies network function dependencies and the orchestrator executes maintenance actions without manual intervention, eliminating human error while maintaining operational control.
Solution Approach 2:
The patent replaces manual mechanical processes with automated computational systems. The AI/ML-based dependency discovery and orchestration system substitutes human teams and manual procedures with intelligent automation, improving both efficiency and reliability.
2Ease of operation
If complex dependencies between network functions are managed manually, then maintenance can be performed, but disruptions occur due to incorrect sequencing
Solution Approach 1:
The system performs preliminary action by automatically discovering and analyzing network function dependencies before maintenance begins. The AI/ML model pre-establishes the correct activation/deactivation sequence, ensuring that maintenance operations proceed without disruptions while simplifying the operational process.
Solution Approach 2:
The orchestration system implements feedback mechanisms to monitor the state of network functions during maintenance. It continuously verifies whether functions are properly activated or deactivated in the correct sequence, adjusting operations as needed to maintain network stability.
3Productivity
If AI/ML-assisted automation is implemented, then maintenance efficiency is improved and errors are reduced, but system complexity increases
Solution Approach 1:
The AI/ML model and orchestration system serve multiple functions: dependency discovery, sequence determination, activation control, and deactivation control. This multi-functionality consolidates what would otherwise require separate systems, managing complexity while maintaining high productivity.
4Measurement precision
If network functions are deactivated individually, then precise control is achieved, but downtime is extended due to sequential operations
Solution Approach 1:
The system performs preliminary dependency analysis to determine the optimal batch sequence for deactivation and activation. By pre-calculating which functions can be safely deactivated together and in what order they should be reactivated, the system minimizes total downtime while maintaining precise control over the maintenance process.
Data Source
AI summary
Aspects of the subject disclosure may include, for example, identifying a set of network functions operative on a core network of a mobile communications system instantiated on a cloud network, identifying functional dependencies among respective network functions of the set of network functions, defining a sequence by which the set of network functions should be made unavailable prior to a maintenance event, wherein the defining the sequence is based on the functional dependencies, and deactivating respective network functions of the set of network functions according to the sequence. Other embodiments are disclosed.