Probabilistic Subscriber Counting in Telecommunication Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for real-time monitoring of telecommunication networks struggle to accurately count distinct subscribers impacted by network issues across multiple time periods due to resource constraints and lack of precision in identifying recurring service failures.
Innovation Solution
A method utilizing a probabilistic counter structure and multi-level final counter aggregation, implemented on a computing device, processes Call Detail Records to estimate the number of distinct subscribers affected by network issues over successive time periods, employing algorithms like Apache Theta Sketch for accurate cardinality estimation and intersection operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-time counting of distinct subscribers is performed using exact methods, then measurement precision is improved, but processing resources and device complexity increase excessively
Solution Approach 1:
The patent transforms the exact counting problem into an estimation problem by changing the parameter from exact count to probabilistic estimate within tolerance. This allows using lighter computational resources while achieving sufficient precision for monitoring purposes.
Solution Approach 2:
The patent employs disposable probabilistic data structures (like HyperLogLog) that can be quickly created and discarded for each counting task, avoiding the need for complex persistent data structures. These lightweight structures provide sufficient precision without requiring extensive processing resources.
2Measurement precision
If exact counting of distinct subscribers is performed, then measurement precision is improved, but processing time increases excessively
Solution Approach 1:
The patent changes the parameter from exact count to estimated count within acceptable tolerance, dramatically reducing processing time while maintaining sufficient precision for real-time monitoring applications.
Solution Approach 2:
The patent performs partial counting using sampling and probabilistic methods rather than exhaustive exact counting. This partial action provides sufficiently accurate results for monitoring purposes while completing the operation within real-time constraints.
3Measurement precision
If the system tracks subscribers affected by network issues over multiple time periods, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the tracking task into separate probabilistic counters for each time period, then combines results using set operations. This segmentation allows complex multi-period analysis to be broken down into simpler independent counting operations that can be processed efficiently.
Solution Approach 2:
The patent uses probabilistic data structures as intermediary representations of subscriber sets across different time periods. These intermediaries enable efficient computation of intersections and unions without requiring direct manipulation of raw subscriber data, reducing overall system complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention concerns a method and a device for estimating a number of distinct subscribers of a telecommunication network impacted by network issues. The method comprises, comprising, for a plurality of N successive counting periods preceding a current time : -determining (40), for each counting period, an estimate of a number of different subscribers impacted by at least one network issue by implementing a probabilistic counter structure, and storing at least one elementary counter in association to said counting period, -aggregating (50) the elementary counters in a multi-level final counter structure, wherein each level of the final counter structure has an associated probabilistic counter structure, the aggregation comprising: - for each elementary counter: computing (58), for at least one level of the multi-level final counter structure, an intersection between said elementary counter and the probabilistic counter structure associated to said level of the multi-level final counter structure, and updating (60-64) the multi-level final counter structure based on the intersection computed.