Telecom Event Inference for Faster Root Cause Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network management systems face challenges in efficiently diagnosing complex network issues due to overwhelming data volumes and the need to navigate multiple UI screens, making it difficult to identify the most probable cause of performance degradation or failures in telecommunication networks.
Innovation Solution
A system that collects data into event objects, applies machine-learning inference functions to determine relationships between event parameters and KPIs, and uses metadata to identify the most probable cause of network issues by reducing data volumes and focusing on anomalous events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional network management systems collect and analyze all network data through multiple hierarchical systems (EMS, NMS, OSS), then comprehensive network monitoring is achieved, but diagnostic time and system complexity increase significantly
Solution Approach 1:
The patent extracts and focuses analysis on only the most relevant parameters and anomalous events rather than processing all network data. The system identifies and isolates key parameters that are most likely to indicate network issues, extracting only this critical subset for detailed analysis while ignoring normal operational data.
Solution Approach 2:
The patent segments the network data analysis into distinct layers: event detection, parameter identification, anomaly detection, and cause determination. Each layer processes specific aspects of the data independently, allowing parallel processing and reducing overall diagnostic time while maintaining comprehensive monitoring.
2Measurement precision
If traditional systems navigate through multiple UI screens to analyze KPI associations and densities, then detailed network diagnostics are obtained, but operator workload and time consumption increase
Solution Approach 1:
The patent implements self-service automation where the system automatically performs parameter identification, anomaly detection, and cause determination without requiring manual navigation through multiple UI screens. The system serves itself by automatically generating diagnostic results and presenting them in a consolidated format, eliminating the need for operators to manually explore multiple interfaces.
Solution Approach 2:
The patent performs preliminary analysis actions automatically before presenting results to operators. The system pre-identifies relevant parameters, pre-detects anomalies, and pre-determines probable causes, so that when operators view the diagnostic information, the heavy analytical work has already been completed in the background.
3Reliability
If network management systems analyze all event parameters and KPI associations, then complete fault coverage is achieved, but data processing volume and resource consumption become overwhelming
Solution Approach 1:
The patent applies local quality by treating different types of events and parameters with different levels of analysis depth. Anomalous events receive intensive analysis with multiple parameters examined, while normal events receive minimal processing. This allows the system to maintain high fault detection coverage for problematic areas while reducing overall data processing volume.
Solution Approach 2:
The patent applies partial action by focusing analysis only on the portion of data that is most likely to contain faults - specifically anomalous events and their associated parameters. Rather than uniformly analyzing all data, the system applies excessive analysis depth only where needed (at anomaly points) and minimal analysis elsewhere, optimizing the balance between coverage and processing volume.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A method performed by a computing system includes collecting information on transactions in a telecommunication system, using the information on transactions to create a plurality of event objects, each of the event objects associated with a telecommunication event, associating each of the event objects with a Key Performance Indicator (KPI), applying the event objects to a plurality of inference functions, each inference functions using the set of parameters as inputs and the KPIs of the event objects as outputs to create a model that infers a relationship between the set of parameters and the KPIs, and analyzing metadata from each of the inference functions to determine which of the set of parameters was used to predict an outcome leading to the KPI.