Radio Network Anomaly Classification via Multi-Domain Data Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In modern cellular systems with disaggregated network components managed by multiple vendors, detecting and determining the root cause of anomalies is challenging due to complexity and difficulty in sharing information between vendors.
Innovation Solution
A system comprising an upper-level controller that analyzes data from both internal and external sources to determine whether an anomaly in a radio network is caused by the network itself or external factors, using correlation matrices and threshold matrices to identify the root cause.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If disaggregated network components are managed by multiple vendors, then system adaptability and versatility are improved, but device complexity and difficulty of detecting anomalies increase
Solution Approach 1:
The patent segments the anomaly detection system into separate functional modules: data collection module, data processing module, correlation analysis module, and classification module. This segmentation allows each module to handle specific tasks independently, reducing overall system complexity while maintaining the ability to handle multi-vendor network environments.
Solution Approach 2:
The patent introduces correlation data as an intermediary between network performance data and external data. This correlation data serves as a mediator that enables the system to identify relationships between different data sources without requiring direct complex integration between all components, thus simplifying the detection process in multi-vendor networks.
2Measurement precision
If multiple data sources are collected and correlated, then anomaly detection precision is improved, but loss of time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-computing correlation matrices between network performance data and external data during normal operation. This allows the system to have ready-to-use correlation information when anomalies occur, eliminating the need for real-time complex correlation calculations and reducing troubleshooting time.
Solution Approach 2:
The system uses feedback from anomaly detection results to refine future detections. By analyzing whether anomalies are correctly identified and adjusting the correlation thresholds and weights accordingly, the system improves precision over time while maintaining efficient detection speeds through learned patterns from previous feedback cycles.
Data Source
AI summary
Generally provided is a radio system that can comprise an upper-level controller that analyzes data comprising first data from a radio network source and second data from an external source that is disposed external to the radio network, and a lower-level controller that is responsive to and provided at a lower level of hierarchy of the radio network than the upper-level controller, where the lower-level controller identifies the first data from the radio network source, and where the upper-level controller classifies, based on the analysis, whether an anomaly, determined to be occurring in a radio network comprising the radio system, is caused by the radio network. The upper-level controller can correlate the first data from the radio network source and the second data from the external source to metrics defining the anomaly.