Device Fingerprinting via Base and Predictor Attributes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing device and app identification technologies face challenges in accurately recognizing returning devices and apps due to changes in device characteristics, such as OS upgrades and carrier changes, leading to high False Accept and False Reject Rates, and are vulnerable to malicious attacks like anonymization and spoofing.

Innovation Solution

The Third Generation fingerprint technology generates base and predictor attributes from device and app data, using machine learning to differentiate between new and returning devices/apps, thereby minimizing False Accept and False Reject Rates by analyzing both stable and unstable information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If device characteristics are collected and transmitted for recognition, then device identification accuracy is improved, but the system becomes vulnerable to malicious attacks and changes in device characteristics cause false rejections

Engineering Contradiction:

Improvedevice identification accuracyVSAvoidrecognition reliability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent segments device identification into two independent components: device fingerprinting (collecting device characteristics) and user behavior analysis (tracking interaction patterns). This segmentation allows the system to maintain reliable recognition even when device characteristics change, as user behavior patterns remain consistent across different devices or app versions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces user behavior patterns as an intermediary layer between device characteristics and identification decisions. Instead of directly relying on device fingerprints, the system uses behavior analysis as a mediator to verify device authenticity, making the identification process more resilient to device changes and attacks.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If device characteristics are monitored for recognition, then returning devices can be identified, but malicious users can anonymize devices to exploit the system

Engineering Contradiction:

Improvedevice recognition efficiencyVSAvoidanonymization attacks

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent implements preliminary action by continuously monitoring and analyzing user behavior patterns before making identification decisions. The system establishes baseline behavior profiles during normal usage, which are then used to detect and prevent anonymization attacks in advance, rather than reacting after attacks succeed.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent employs feedback mechanisms where user behavior data is continuously collected, analyzed, and used to refine identification accuracy. The system learns from ongoing interactions, adjusting its recognition algorithms to detect anonymization attempts and improve differentiation between legitimate returning devices and malicious actors.

Inventive Principle:

Principle #23Feedback

3Reliability

If device characteristics are used for identification, then authorized devices can be recognized, but spoofing attacks can lead to false acceptances

Engineering Contradiction:

Improveauthorized device recognitionVSAvoiddevice differentiation accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent changes the parameters used for identification from static device characteristics to dynamic user behavior patterns. By monitoring how users interact with the device over time (typing patterns, navigation behaviors, usage timing), the system creates a more difficult-to-spoof identification mechanism that maintains high accuracy in distinguishing authorized devices from impostors.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent transitions from static device fingerprinting to dynamic behavior analysis. User behavior patterns evolve and adapt over time, making the identification system more robust against spoofing attacks. The dynamic nature of behavioral biometrics requires attackers to not only copy device characteristics but also replicate complex usage patterns, significantly increasing the difficulty of successful spoofing.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11403563B2Systems and methods for facilitating recognition of a device and/or an instance of an app invoked on a device

Publication Date: 2022.08.02 ACCERTIFY INC

AI summary

A system of classifying devices and/or app instances a new or returning divides attributes generated from observations received from an uncharacterized device/software app into base-fingerprint attributes and predictor attributes, where the two kinds of attributes have different longevities. Predictor attribute tuples from attribute tuples having the same base fingerprint as the base fingerprint corresponding to the uncharacterized device/app, and the predictor attribute tuple corresponding to the uncharacterized device/app are analyzed using a machine learned predictor function to obtain a final fingerprint. Machine learning techniques such as logistic regression, support vector machine, and artificial neural network can provide a predictor function that can decrease the conflict rate of the final fingerprint and, hence, the utility thereof, without significantly affecting the accuracy of classification.