System State Graph Models for Network Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing OSS solutions for managing complex network systems, such as communications networks, are stateful and lack the capability for continuous action and effect capturing, proactive decision-making, and efficient online updates, especially with the increased complexity of 5G networks, leading to unreliable models due to divergent cell properties.
Innovation Solution
A method that generates central and updated system state graph models to represent operational units' properties as probabilities of transitions between states, with distance measures determining the need for new central models to manage divergent subsets of operational units, allowing for real-time, manageable, and accurate decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single central system state graph model is used to represent all operational units, then the model complexity is reduced and data processing is simplified, but the model becomes unreliable when operational units have divergent properties
Solution Approach 1:
The patent divides the operational units into multiple clusters based on their properties and generates separate central system state graph models for each cluster. This segmentation allows each model to represent units with similar characteristics accurately, improving reliability while managing complexity through organized grouping rather than a single monolithic model.
Solution Approach 2:
The patent implements dynamic model generation where the system continuously monitors operational units and creates or updates central models based on real-time data. When units diverge from existing clusters, new clusters and models are dynamically created. This dynamic adaptation maintains model reliability without requiring static, overly complex structures.
2Measurement precision
If heavy data pre-processing and complex online updates are performed to maintain accurate models, then decision-making accuracy is improved, but the computational complexity and processing time increase significantly
Solution Approach 1:
The patent performs preliminary actions by pre-defining clusters of operational units based on their properties before making decisions. This pre-organization allows the system to use simpler, pre-computed models for each cluster rather than performing complex real-time analysis, improving decision accuracy while reducing computational complexity during execution.
Solution Approach 2:
The patent creates simplified representations (copies) of operational units in the form of central system state graph models for each cluster. These copies capture essential characteristics without requiring full complexity of individual units, enabling accurate decision-making through the simplified models while reducing processing requirements compared to analyzing every unit in detail.
3Loss of information
If machine learning algorithms are used to extract insights from raw data, then insights can be obtained, but the pool of suitable algorithms with efficient online updates is limited and model updates become less intuitive
Solution Approach 1:
The patent replaces complex machine learning algorithms with a rule-based clustering approach that groups operational units by their properties and generates corresponding central models. This substitution maintains the ability to extract meaningful insights from data while using simpler, more interpretable methods that are easier to update and modify, avoiding the black-box nature of many ML algorithms.
Data Source
AI summary
The properties of a plurality of operational units are estimated by generating a central system state graph model representing the properties of the plurality of operational units as probabilities of transitions between states for the plurality of operational units, where the states represent operational data. Then a respective updated system state graph model is generated for each of the plurality of operational units, based on the central system state graph model and based on new operational data for the respective operational unit. A distance measure is determined between the respective updated system state graph models. If the distance measure fulfils a divergence criterion, a plurality of new central system state graph models are generated, each representing the properties of a respective subset of the plurality of operational units as the probabilities of transitions between states for the respective subset of the plurality of operational units.