Adaptive Event-Based Network Analytics for 5G
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network analytics systems face significant hardware resource challenges due to the large amount of data processing required for event-based monitoring in 5G mobile networks, particularly in detecting service quality degradation and network issues, which leads to high processing and storage demands.
Innovation Solution
An adaptive event-based network analytics system that dynamically filters event data sources and selectively monitors a subset of subscribers and events, using location and activity models to identify statistically relevant samples for symptom detection, thereby reducing hardware requirements by focusing on a representative subset of the subscriber base.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full event-based monitoring is implemented for all subscribers in 5G networks, then service quality detection capability is improved, but hardware resource requirements increase significantly
Solution Approach 1:
The patent segments the subscriber base into different groups based on service quality relevance and monitoring needs. Instead of monitoring all subscribers uniformly, the system divides them into segments where only those with detected service quality issues undergo detailed event-based analysis, while others receive simplified monitoring. This segmentation reduces the overall data processing load while maintaining detection capability for problematic cases.
Solution Approach 2:
The patent applies partial action by implementing event-based monitoring selectively rather than universally. The system performs comprehensive event collection and correlation only for subscribers where service quality degradation is detected or suspected, while using lighter monitoring approaches for the majority of subscribers. This partial application of full monitoring capabilities reduces hardware resource demands while preserving the ability to detect and analyze service quality issues when they occur.
2Measurement precision
If comprehensive event data collection is performed for all subscribers, then troubleshooting accuracy is improved, but processing time increases
Solution Approach 1:
The patent implements preliminary action by continuously collecting and pre-processing event data in the background before service quality issues manifest. The system maintains pre-computed event correlations and subscriber profiles that are ready for immediate analysis when degradation is detected. This preliminary preparation reduces the processing time required when actual troubleshooting is needed, as the heavy lifting of data collection and initial correlation has already been performed.
Solution Approach 2:
The patent applies dynamics by adapting the level of monitoring detail based on real-time network conditions and subscriber behavior. The system dynamically adjusts between simplified and comprehensive monitoring modes, collecting full event data only when service quality issues are detected. This dynamic adjustment optimizes processing time by avoiding unnecessary comprehensive analysis during normal operating conditions while ensuring accurate troubleshooting when problems occur.
3Adaptability or versatility
If event-based analytics is applied to all subscribers, then service quality monitoring coverage is improved, but system complexity increases
Solution Approach 1:
The patent implements universality by designing a multi-functional analytics system that can operate in multiple modes depending on the situation. The same core event collection infrastructure supports both simplified monitoring for the majority of subscribers and comprehensive event-based analysis for those with service quality issues. This universal platform provides broad monitoring coverage while managing complexity through unified architecture that adapts its processing depth based on subscriber needs.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An event-based network analytics system is provided for monitoring KPIs. The event-based analytics system implements an adaptive method and analytical system is provided for reducing the required hardware resources of event-based network analytics system by monitoring a subset of KPIs, which are used for symptom detections, and dynamic filtering in the different event data sources in order to ensure the statistically required sampled for calculating data rate for symptom detection. In addition, the monitoring of the limited set of events is only performed for an adaptively chosen small subset of the total subscriber base that is estimated to cover the geographical entirety of the mobile network in a statistically reliable way.