Authentication Filters for Computing Device Signal Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication networks face challenges in securely authenticating computing devices, particularly in distinguishing between in-distribution and out-of-distribution devices, which can lead to unauthorized access and compromised network security.
Innovation Solution
A computer-implemented method and system that utilizes a trained authentication model to classify computing devices based on signal data, generating authentication filters to determine device authenticity, and adjusts network access accordingly, employing machine-learned models like generative neural networks and autoencoders to process physical signal samples and generate device fingerprints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine-learned models are used to classify computing devices based on signal data, then authentication accuracy is improved, but computational complexity and resource requirements increase
Solution Approach 1:
The authentication system is divided into multiple functional components: signal collection module, machine-learned model classification module, and authentication filter generation module. This segmentation allows each component to process specific tasks independently, improving overall efficiency while maintaining high authentication accuracy through specialized processing at each stage.
Solution Approach 2:
The system generates synthetic signal data that replicates the characteristics of legitimate device signals. These synthetic copies are used to train authentication filters and machine-learned models, enabling accurate authentication without requiring complex real-time analysis of every incoming signal, thus reducing computational burden while maintaining precision.
2Speed
If authentication filters are generated and stored for classification, then authentication speed is improved, but storage requirements and system complexity increase
Solution Approach 1:
Authentication filters are pre-generated and stored during system initialization or offline processing. By performing this action in advance, the system can quickly match incoming device signals against pre-computed filters during actual authentication operations, significantly improving authentication speed without requiring complex real-time computations.
Solution Approach 2:
The system introduces an intermediary authentication filter layer between the raw signal data and the final authentication decision. This intermediary component simplifies the authentication process by translating complex signal analysis into simpler filter-matching operations, reducing overall system complexity while maintaining fast authentication capability.
3Reliability
If network access is adjusted based on authentication classification, then security is improved, but system response time increases
Solution Approach 1:
Authentication decisions and network access policies are pre-established and stored. When a device attempts to access the network, the system quickly matches the device against pre-computed authentication filters and retrieves the corresponding access decisions from pre-stored policy data, eliminating the need for complex real-time policy evaluation and significantly reducing response time while maintaining high security.
Data Source
AI summary
Systems and methods for computing device authentication can involve template matching with an authentication filter. Authentication filters can be generated by using an authentication model to determine or generate templates for authenticated classes. For example, simulated signal data sets can be input into the authentication model until an authenticated class classification is output. The successful simulated signal data set may then be used to generate or update an authentication filter that uses the template for template matching authentication classification.