Traffic Prediction Model for Dynamic Warning Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network traffic monitoring methods rely on fixed thresholds, leading to inaccurate warnings due to varying network usage patterns, resulting in both false alarms and missed alerts.
Innovation Solution
A method and apparatus that utilize a pre-trained traffic prediction model, incorporating convolutional neural networks, residual networks, and fully connected layers, to accurately predict network traffic by analyzing real-time data and sending warnings when differences between predicted and actual traffic exceed a set threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a fixed traffic threshold is used for monitoring, then the monitoring system is simple to operate, but the warning accuracy deteriorates due to varying network usage patterns
Solution Approach 1:
The patent transforms the static fixed threshold into a dynamic prediction-based threshold. The system uses historical traffic data to train a prediction model that dynamically adjusts the threshold according to actual network usage patterns, thereby maintaining both ease of operation and high warning accuracy.
Solution Approach 2:
The patent changes the parameter from a fixed threshold value to a dynamically calculated predicted traffic value. By using machine learning models to predict future traffic based on historical patterns, the system adapts the threshold parameter to match actual network conditions, resolving the contradiction between simplicity and accuracy.
2Device complexity
If a fixed traffic threshold is used for monitoring, then the system complexity is low, but false alarms increase due to inaccurate warnings
Solution Approach 1:
The patent applies preliminary action by training the prediction model in advance using historical traffic data before actual monitoring begins. This pre-training phase allows the system to learn normal traffic patterns and establish accurate prediction baselines, reducing false alarms during operational monitoring without adding significant complexity.
Solution Approach 2:
The system implements feedback by continuously comparing actual traffic against predicted traffic and using the differences to refine predictions. This feedback mechanism improves warning reliability over time while maintaining relatively simple system architecture through iterative learning.
3Ease of manufacture
If a fixed traffic threshold is used, then the monitoring approach is easy to implement, but missed alerts occur due to not reflecting actual usage trends
Solution Approach 1:
The patent enables the monitoring system to serve itself by automatically learning traffic patterns from historical data without requiring manual threshold adjustment. The prediction model self-adapts to changing network usage trends, maintaining high detection precision while keeping implementation simple through automated learning processes.
Data Source
AI summary
Embodiments of the present disclosure relate to a method and apparatus for warning. A method may include: importing a real-time collected traffic data sequence into a pre-trained traffic prediction model, to obtain predicted traffic data, the traffic prediction model being used to calculate and obtain the predicted traffic data according to the real-time collected traffic data sequence; and sending, in response to a difference between the predicted traffic data and actual traffic data corresponding to time being greater than a set threshold, a warning signal.