Mobile Network Policy Validation via Distributed Self-Service
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for implementing policies in mobile networks are costly and time-consuming due to the need for centralized creation and deployment of policy bases by expert personnel, exacerbated by the complexity introduced by machine learning in cognitive networks, which makes it difficult to ensure consistency between human-created and automatically generated policies.
Innovation Solution
A method where human-created policies are validated using machine learning, with a validation algorithm sent to each network element to check policy consistency based on its policy base and system state, allowing for decentralized validation and reducing the need for extensive expert involvement, and enabling the use of machine learning to augment policy sets while ensuring consistency through a hierarchical validation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If policies are created and validated centrally in the operations and support system, then policy consistency can be ensured, but the process becomes costly and time-consuming requiring extensive expert personnel
Solution Approach 1:
The patent divides the centralized policy validation process into distributed validation across multiple network elements. Each network element validates policies locally using received validation algorithms, splitting the monolithic validation task into parallel distributed operations that reduce central system burden and accelerate policy deployment
Solution Approach 2:
Network elements perform self-validation of policies using validation algorithms received from the operations and support system. This self-service approach eliminates the need for extensive expert personnel to manually validate each policy centrally, reducing costs and time while maintaining consistency through automated validation logic
2Productivity
If machine learning is used to automatically generate policies, then policy creation efficiency improves, but ensuring consistency between human-created and automatically generated policies becomes difficult
Solution Approach 1:
The validation algorithm acts as an intermediary between human-created policies and machine learning-generated policies. This intermediary validation layer ensures that both types of policies meet consistency requirements before deployment, resolving the conflict between automated policy generation efficiency and policy consistency reliability
Solution Approach 2:
The system implements feedback mechanisms where validation results from network elements are reported back to the operations and support system. This feedback loop allows continuous verification of policy consistency between human-created and machine-generated policies, enabling automated policy creation while maintaining reliability through iterative validation
3Reliability
If validation is performed centrally for all network elements, then comprehensive policy validation is achieved, but operational effort and time requirements increase significantly
Solution Approach 1:
The validation process is segmented and distributed to individual network elements that validate their own policies locally. This segmentation transforms a time-consuming centralized validation process into parallel distributed validation operations, maintaining comprehensive coverage while dramatically reducing total validation time
Solution Approach 2:
Each network element performs self-validation using received validation algorithms, eliminating the need for centralized validation of every network element. This self-service approach maintains validation completeness while reducing operational effort and time requirements by enabling autonomous validation at the network element level
Data Source
Figure 1~2
Figure 3~4
AI summary
It is described a method for implementing policies in a mobile network (1) with at least one operations and support system (2) and a number of network elements (30..33) connected thereto. According to said method, a set of policies is validated separately for each network element (30..33), for which said policy set is relevant. Furthermore, a mobile network (1) and a network element (30..33) for performing said method are disclosed.