Passive Biometric Authentication Normalization Across Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing passive biometric authentication systems face challenges in accurately authenticating users due to variations in device properties and usage modes, leading to increased risks of false positives and false negatives.

Innovation Solution

The proposed system normalizes user inputs by defining abstracted device classes that account for device properties and usage modes, allowing for effective authentication across different devices and usage scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If passive biometric authentication is implemented without normalization, then user authentication can be performed across different devices, but the accuracy decreases due to device property variations leading to false positives and false negatives

Engineering Contradiction:
Improvecross-device authentication capabilityVSAvoidauthentication accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by normalizing biometric input features based on device-specific parameters. Different device properties (screen size, sensor characteristics, usage patterns) are used as parameters to transform and standardize the biometric data, allowing accurate authentication across diverse devices while accounting for their unique characteristics

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces an intermediary normalization layer between the raw biometric inputs and the authentication decision. This intermediary process adjusts the biometric features based on device properties, acting as a mediator that bridges the gap between device variability and authentication accuracy, thereby resolving the contradiction between cross-device adaptability and measurement precision

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If device-specific biometric models are created for each device, then authentication accuracy improves, but system complexity increases

Engineering Contradiction:
Improveauthentication accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements universality by creating a unified authentication framework that works across multiple devices. Instead of maintaining separate authentication systems for each device, a single normalized model can serve multiple device types, reducing system complexity while maintaining accuracy through the normalization of device-specific variations

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Reliability

If normalization based on device classes is implemented, then false authentication risk decreases, but processing time increases due to device class identification and normalization steps

Engineering Contradiction:
Improveauthentication reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-defining device classes and their associated normalization parameters before actual authentication occurs. During runtime, the system quickly identifies which pre-defined device class the current device belongs to and applies the corresponding normalization, avoiding the need for complex real-time analysis and reducing processing time while maintaining high reliability

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3992823B1System and method for user authentication based on passive biometrics across different devices
Publication Date: 2025.05.14 BLACKBERRY LTD
  • EP3992823B1 patent drawingFigure 1
  • EP3992823B1 patent drawingFigure 2
  • EP3992823B1 patent drawingFigure 3

AI summary

A computer-implemented method is disclosed. The method includes: receiving, via an input interface associated with an electronic device, a first user input; identifying a first device class based on a current usage mode of the electronic device and a first set of device property values associated with the current usage mode of the electronic device; obtaining normalized values of input features associated with the first user input, the input feature values being normalized based on representative device property values that are defined for the first device class; and performing authentication of a user of the electronic device based on the normalized input feature values associated with the first user input.