Mobile Network Problem Cell Detection Through KPI Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying problem cells in mobile communication networks are inaccurate and reliant on human judgment, failing to accurately pinpoint cells causing network anomalies due to varying traffic volumes and subjective threshold settings.
Innovation Solution
A method to determine anomaly contribution degrees of each cell by correlating key performance indicators with network anomalies, using dynamic threshold detection and automated analysis to identify cells causing subnet issues, thereby improving accuracy and reducing human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual judgment and fixed threshold methods are used to identify problem cells, then the process is simple to implement, but the identification accuracy is low and reliant on human judgment
Solution Approach 1:
The patent replaces manual judgment and fixed threshold settings with an automated anomaly contribution degree calculation system. The system uses dynamic threshold detection and automated analysis to identify problem cells, substituting human judgment with algorithmic processing that calculates contribution degrees based on key performance indicators and their correlation with network anomalies.
Solution Approach 2:
The patent transforms the identification approach by changing from fixed thresholds to dynamic anomaly contribution degrees. The system calculates contribution degrees using multiple parameters including key performance indicators, traffic volumes, and correlation coefficients, allowing the thresholds to adapt dynamically rather than remaining static.
2Measurement precision
If fixed threshold methods are used for cell identification, then the method is easy to implement, but it fails to account for varying traffic volumes and leads to inaccurate identification
Solution Approach 1:
The patent introduces dynamic threshold detection that adapts to varying traffic volumes and network conditions. Instead of using fixed thresholds, the system dynamically adjusts thresholds based on actual network data, traffic patterns, and anomaly detection, making the identification process responsive to changing conditions while maintaining accuracy.
Solution Approach 2:
The system performs preliminary analysis by calculating anomaly contribution degrees for all cells before final identification. This preliminary calculation of contribution degrees based on key performance indicators and correlation analysis prepares the data structure needed for accurate problem cell identification, accounting for varying traffic volumes in advance.
3Extent of automation
If automated analysis with dynamic thresholds is implemented, then identification accuracy improves and human intervention is reduced, but the system complexity increases
Solution Approach 1:
The system implements self-service through automated anomaly detection and contribution degree calculation. The network management system automatically collects key performance indicators, calculates anomaly contribution degrees, identifies problem cells, and generates optimization recommendations without requiring manual intervention, making the complex automated process self-executing.
Solution Approach 2:
The system incorporates feedback mechanisms where the results of anomaly contribution degree calculations are fed back into the identification process. The system continuously monitors key performance indicators, recalculates contribution degrees based on new data, and adjusts its identification criteria, creating a closed-loop automated system that refines its accuracy over time.
Data Source
AI summary
A method for identifying a problem cell, an electronic device and a computer-readable medium Said method comprises: determining the anomaly contribution degree of each cell in a subnet (S101); and determining at least one cell as a problem cell according to the anomaly contribution degree of each cell in the subnet (S102); the anomaly contribution degree of each cell refers to, when the key performance indicator of the subnet is anomalous, the degree of correlation between the key performance indicator of the cell and the anomaly; the key performance indicator of the subnet being anomalous means that the key performance indicator of the subnet exceeds a first threshold range; the key performance indicator of the subnet is determined according to parameter statistics of cells therein; and the key performance indicator of the cell is determined according to parameter statistics therein. Disclosed are a method for identifying a problem cell, an electronic device, and a non-transitory computer-readable medium. The method includes: determining an anomaly contribution degree of each cell in a subnet; and determining at least one cell as a problem cell according to the anomaly contribution degree of each cell in the subnet.