Dynamic Fingerprinting via Fuzzy Hash Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing device fingerprinting systems face challenges in balancing data privacy and accuracy, as they require extensive data points that may change over time, leading to authentication issues and privacy concerns.
Innovation Solution
The implementation of dynamic and private security fingerprinting using fuzzy hashes, where a server computer generates a fingerprint value based on user data, compares it to a baseline, and updates the baseline if the similarity score exceeds a threshold, allowing for authentication without exact matching and preserving user privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If more data points are used to generate the device fingerprint, then the accuracy and reliability of authentication is improved, but user privacy concerns increase and data collection becomes more complex
Solution Approach 1:
The patent extracts only the essential characteristics needed for authentication from the device data, rather than collecting and storing all raw data points. By using fuzzy hashing, the system extracts a condensed representation that maintains authentication accuracy while minimizing privacy intrusion through selective data extraction.
Solution Approach 2:
The patent transforms device data parameters through fuzzy hashing, converting detailed device characteristics into a condensed fingerprint representation. This parameter transformation maintains the unique identifying features necessary for authentication while reducing the amount of sensitive information that needs to be stored and transmitted.
2Measurement precision
If more data points are collected for the fingerprint, then the fingerprint accuracy is improved, but the device complexity and data management burden increase
Solution Approach 1:
The system extracts only the necessary data points required for accurate fingerprinting, eliminating redundant data collection. By focusing on essential device characteristics and using fuzzy hashing to process them, the system achieves high fingerprint accuracy without the complexity of managing extensive datasets.
Solution Approach 2:
The patent performs preliminary fuzzy hashing of device data during the fingerprinting process, transforming raw data into a condensed representation before storage. This preliminary processing reduces the complexity of data management while maintaining the precision needed for accurate authentication.
3Reliability
If traditional exact matching is used for fingerprint comparison, then authentication security is improved, but the system becomes less adaptable to legitimate device changes
Solution Approach 1:
The patent implements dynamic fingerprint comparison using fuzzy matching, allowing the authentication system to adapt to legitimate device changes. The fuzzy matching mechanism dynamically adjusts the matching criteria, maintaining security by requiring substantial similarity while accommodating normal device modifications such as OS updates or configuration changes.
Solution Approach 2:
The system changes the matching parameter from exact equality to fuzzy similarity, allowing authentication to succeed when device fingerprints are substantially similar but not identical. This parameter change enables the system to maintain security while being adaptable to legitimate device evolution.
Data Source
AI summary
Techniques for securely generating and using a “fingerprint” for authentication. A server computer receives a first data set from a user device (including a first fuzzy hash of first user data on the user device). The server computer generates a first fingerprint value based on the first data set. The server computer detects an event corresponding to a user in association with the user device. The server computer identifies a baseline fingerprint value (generated based on a baseline fuzzy hash of user data on the user device). The server computer compares the first fingerprint value to the baseline fingerprint value to generate a similarity score. The server computer may determine that the similarity score exceeds a threshold value but does not represent an exact match, and, based on the similarity score, authenticate the user and update the baseline fingerprint value based on the first fingerprint value.