Edge Cyberattack Detection Using Supervised Random Forest
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting cyberattacks, such as DDoS and brute force attacks, on electronic devices are inadequate as they often rely on external data centers or cloud computing, leading to inefficiencies and potential loss of critical information.
Innovation Solution
Implementing a supervised machine-learning model, specifically a Random Forest model, on the electronic device itself for real-time data classification and analysis, allowing for local detection and mitigation of cyberattacks without relying on external platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If external data centers or cloud computing are used for cyberattack detection, then computational resources and detection capabilities are enhanced, but system complexity increases and critical information may be lost
Solution Approach 1:
The patent extracts the machine learning model and detection algorithms from external cloud computing environments and implements them directly on edge devices. This allows the device to perform cyberattack detection locally using its own processing power, eliminating the need for complex external infrastructure while maintaining detection capabilities
Solution Approach 2:
The edge device performs self-service by executing the machine learning model locally to detect cyberattacks without requiring external processing. The device autonomously analyzes its own operational data, classifies threats, and responds to attacks independently, reducing system complexity while maintaining reliable detection
2Loss of time
If data is collected and analyzed locally on the electronic device, then user data privacy is maintained and response time is reduced, but the device requires sufficient processing power and memory
Solution Approach 1:
The patent applies preliminary action by pre-training the machine learning model on a comprehensive dataset of cyberattack patterns before deployment. The model is trained in advance to recognize various attack types, enabling the device to perform rapid local analysis and classification of incoming data without requiring excessive processing power during actual attack detection
Solution Approach 2:
The system optimizes parameters such as the number of decision trees in the Random Forest model, feature selection, and classification thresholds to achieve effective cyberattack detection with limited computational resources. By carefully tuning these parameters, the device can maintain fast response times while operating within its hardware capabilities
3Measurement precision
If a supervised machine-learning model is deployed on the electronic device, then detection accuracy and timeliness are improved, but the device needs to manage and update the model
Solution Approach 1:
The patent implements feedback mechanisms where the device continuously monitors its detection performance and compares results against known attack patterns. This feedback loop enables the system to refine its classifications, reduce false positives, and maintain high detection accuracy while operating autonomously without complex external management
Data Source
AI summary
A method for detecting a cyberattack on an electronic device is provided. The method is performed by the electronic device itself. The method includes collecting data at the electronic device. Further, the method includes classifying the collected data as regular data or malicious data using a supervised machine-learning model for the cyberattack. The method additionally includes determining whether the electronic device is under the cyberattack based on the classification of the collected data.