ML-Based Network Service Attack Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional solutions fail to effectively detect and mitigate service degradation in network services, leading to potential attacks that consume major resources and degrade network performance.
Innovation Solution
A computer-implemented method and system using Machine Learning (ML) models to collect operational data, detect degradation, and temporarily disconnect suspected connections, trained to predict the impact of operational factors on resource utilization, thereby mitigating potential network service attacks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional rule-based methods are used to detect service degradation, then the system structure remains simple, but the detection accuracy and ability to identify unknown attack patterns is insufficient
Solution Approach 1:
The patent replaces conventional rule-based detection methods with machine learning models that automatically learn impact patterns from operational data. The ML models substitute manual rule configuration with automated pattern recognition, achieving superior detection accuracy for both known and unknown attack patterns while maintaining manageable system complexity through standardized model deployment.
2Adaptability or versatility
If Machine Learning models are deployed to detect service degradation, then the detection accuracy and adaptability improve, but the computational resources and system complexity increase
Solution Approach 1:
The patent applies machine learning models to predict future resource utilization trends before actual service degradation occurs. By analyzing current operational factors and learned impact patterns, the system proactively identifies connections likely to cause degradation and disconnects them in advance, preventing service disruption before it happens.
Solution Approach 2:
The system continuously collects operational data, applies ML models to predict impact patterns, disconnects suspected malicious connections, and monitors subsequent service performance. This closed-loop feedback mechanism allows the system to learn from actual outcomes and improve detection accuracy over time, enhancing adaptability while managing complexity through iterative optimization.
3Reliability
If traditional rule-based methods are used, then the response time is fast for known patterns, but the system cannot detect previously unknown attack patterns
Solution Approach 1:
The patent replaces rule-based detection with machine learning models that automatically recognize attack patterns based on learned impact relationships. This substitution enables the system to detect both known and previously unknown attack patterns by identifying anomalous operational factor combinations that correlate with service degradation, maintaining fast response times through efficient model inference.
Data Source
AI summary
Disclosed herein are systems and methods for automatically mitigating potential network services attacks based on service usage patterns learned using Machine Learning (ML) comprising, collecting operational data indicative of resource utilization of one or more network services serving a plurality of connections and of a plurality of operational factors of the plurality of connections, detecting degradation of the network service(s) based on analysis of the operational data, applying trained ML model(s) to the operational data in order to identify negative operational factor(s) of one or more suspected connections to the network service estimated to induce the degradation where the one or more ML model is trained to predict an impact pattern induced by each of a plurality of operational factors on the resource utilization of the one or more network services, and disconnecting, at least temporarily, the suspected connection(s) from the network service(s).