Telemetry Anomaly Detection via Distributional Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data processing approaches struggle to efficiently store and analyze large volumes of telemetry data, leading to accuracy issues and error-prone conclusions due to limited analysis.
Innovation Solution
The method involves generating reference data distributions from historical telemetry data using artificial intelligence techniques and comparing them to current data distributions for anomaly detection, using distributional distance determinations to identify anomalies and trigger automated actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional data processing approaches are used to store and analyze telemetry data, then storage costs and resource requirements increase significantly, but analysis accuracy deteriorates due to limited data analysis
Solution Approach 1:
The patent extracts only the essential distributional characteristics from telemetry data using AI techniques, rather than storing and analyzing all raw data. This extraction process captures the meaningful patterns while discarding redundant information, thereby improving analysis accuracy without requiring proportional increases in storage capacity
Solution Approach 2:
The patent transforms raw telemetry data into distributional parameters and characteristics through AI processing. By changing the data representation from raw values to distributional parameters, the system achieves more accurate anomaly detection while reducing the effective data volume that needs to be stored and processed
2Loss of information
If large volumes of telemetry data are stored using conventional approaches, then storage capacity requirements increase, but the ability to perform meaningful analysis remains limited
Solution Approach 1:
The patent replaces conventional mechanical storage and processing systems with AI-based distributional analysis. Instead of relying on brute-force storage and processing of all data, the system uses intelligent algorithms to extract distributional characteristics, thereby improving information extraction capability while reducing storage system complexity
3Reliability
If portions of telemetry data are analyzed due to resource constraints, then processing resources are reduced, but analysis reliability deteriorates due to error-prone conclusions
Solution Approach 1:
The patent performs preliminary AI-based processing to extract distributional characteristics before anomaly detection. This preliminary action prepares the data in a form that enables reliable analysis with reduced resource consumption during the actual anomaly detection process, thereby improving conclusion reliability while lowering processing resource requirements
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for detecting anomalies in device telemetry data using distributional distance determinations are provided herein. An example computer-implemented method includes generating at least one reference data distribution for at least one telemetry data-related metric by processing historical telemetry data derived from devices using artificial intelligence techniques; generating, for at least one device, at least one data distribution for the at least one telemetry data-related metric by processing telemetry data derived from the at least one device using the artificial intelligence techniques; determining one or more distributional distance values by comparing at least a portion of the at least one data distribution to at least a portion of the at least one reference data distribution; identifying one or more anomalies based on the one or more distributional distance values; and performing automated actions based on the one or more identified anomalies.