ML Network Connectivity Monitoring via Edge Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems fail to effectively monitor network connectivity across a wide variety of deployed devices, leading to undetected and unaddressed connectivity issues due to the complexity of devices, communication technologies, and geographic distributions.
Innovation Solution
A method using machine learning models to analyze historical and current connectivity records, selecting appropriate model types based on data volume and quality, and generating forecasted connectivity records to identify anomalies and initiate remedial actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional monitoring systems are used to track network connectivity across diverse devices and regions, then device coverage is extensive, but detection precision deteriorates due to system complexity and inability to handle variability
Solution Approach 1:
The patent segments the monitoring task by deploying distributed edge computing nodes across different geographic regions and device types. Each node independently processes connectivity data for its local segment, transforming a single complex centralized system into multiple simpler distributed units. This segmentation enables precise local detection while reducing overall system complexity through modular architecture.
Solution Approach 2:
The patent introduces machine learning models as intermediary components between raw connectivity data and detection decisions. These ML models act as intelligent mediators that process complex multi-source data (device logs, network telemetry, geographic information) and translate them into actionable connectivity assessments, thereby improving detection precision without requiring the monitoring system itself to become overly complex.
2Loss of information
If distributed devices transmit frequent status updates to central system, then monitoring coverage is comprehensive, but data transmission time increases due to network connectivity issues
Solution Approach 1:
The patent implements preliminary action by pre-configuring edge computing nodes with machine learning models and processing capabilities before connectivity issues occur. These nodes continuously pre-process connectivity data locally, maintaining ready-to-analyze processed information that can be quickly transmitted or acted upon even when network connectivity is interrupted, thus preventing data loss without requiring frequent real-time transmissions.
Solution Approach 2:
The patent extracts the data processing function from the centralized system and relocates it to distributed edge nodes. By taking out the computational burden from the central system and placing it at the network edge where data is generated, the system achieves comprehensive monitoring coverage while minimizing data transmission requirements and associated time delays.
3Measurement precision
If centralized system processes all connectivity data, then analysis accuracy is high, but processing speed deteriorates due to data volume and transmission delays
Solution Approach 1:
The patent segments the centralized data processing function into distributed processing units at the network edge. Each edge node independently analyzes connectivity data for its local region using embedded machine learning models, enabling parallel processing across multiple nodes. This segmentation simultaneously improves processing speed through concurrency and maintains analysis accuracy by preserving detailed local data context that would be lost in centralized aggregation.
Solution Approach 2:
The patent transitions from a single-dimensional centralized processing architecture to a multi-dimensional distributed architecture. By adding the spatial dimension of distribution across multiple geographic locations and the functional dimension of localized intelligence, the system achieves both high-speed parallel processing and high-accuracy local analysis without the bottlenecks of centralized processing.
Data Source
AI summary
Techniques for monitoring network connectivity using machine learning are provided. A plurality of historical connectivity records is received, and a first machine learning model type, of a plurality of machine learning model types, is selected based on the plurality of historical connectivity records. A machine learning model, of the first machine learning model type, is trained based on the plurality of historical connectivity records, where the machine learning model learns to generate forecasted connectivity records based on the training.