ML Model Predicts Network Events Using Temporal Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional network systems rely on manual and reactive processes to adjust Interior Gateway Protocol (IGP) metrics, which are time-consuming, prone to errors, and often fail to anticipate issues before noticeable delays occur, leading to inefficient network management and potential service disruptions.
Innovation Solution
The implementation of a system that uses machine learning models to analyze Performance Monitoring data from optical networks, predicting impending changes to IGP metrics by identifying underlying issues and leveraging temporal and spatial correlations, allowing for proactive adjustments to prevent network disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual processes are used to adjust IGP metrics reactively, then network operators can respond to detected issues, but the process is time-consuming and delays network recovery
Solution Approach 1:
The system performs preliminary actions by predicting link failures before they occur using machine learning models that analyze historical Performance Monitoring data. This allows network operators to proactively adjust IGP metrics before actual failures happen, eliminating the reactive delay inherent in manual processes.
Solution Approach 2:
The patent replaces the manual mechanical process of IGP metric adjustment with an automated system using machine learning models and algorithms. The system automatically analyzes PM data, predicts failures, and can automatically adjust IGP metrics without human intervention, significantly reducing response time while maintaining or improving reliability.
2Measurement precision
If manual analysis of Performance Monitoring data is performed, then network operators can identify problematic links, but the process is complex and prone to errors
Solution Approach 1:
The system replaces manual analysis with automated machine learning models that process Performance Monitoring data. These models use sophisticated algorithms to identify patterns and predict failures with high accuracy, eliminating human error while reducing operational complexity through automation.
Solution Approach 2:
The system creates a virtual model or copy of the network behavior by training machine learning models on historical PM data. This digital twin allows the system to simulate and predict future network states without requiring manual analysis of actual network conditions, improving detection accuracy while simplifying the operational process.
3Ease of manufacture
If static rules are used to configure IGP metrics, then configuration is simplified, but the rules may not be optimized and are hard to change
Solution Approach 1:
The system transitions from static IGP metric configurations to dynamic, adaptive configurations driven by machine learning predictions. The system continuously learns from new data and automatically adjusts IGP metrics in response to predicted failures, providing both ease of operation through automation and adaptability to changing network conditions.
Solution Approach 2:
The system implements a feedback loop where machine learning models continuously analyze Performance Monitoring data, predict potential failures, and adjust IGP metrics accordingly. This closed-loop system automatically adapts to new network conditions and learned patterns, maintaining optimal performance while simplifying configuration management through automated decision-making.
Data Source
AI summary
Systems and methods for predicting network events are provided. A process, according to one implementation, includes the step of receiving a time-series dataset having a sequence of datapoints each including a set of Performance Monitoring (PM) parameters of a network. The process also includes the step of applying a subset of the sequence of datapoints to a Machine Learning (ML) model having a classification function and an encoding/decoding function. In addition, the process includes the step of allowing the ML model to leverage temporal-based correlations among the datapoints of the subset to predict an event associated with the network.