Passive Sensor Authentication Using Accelerometer Feature Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user authentication systems in computing environments face challenges such as password vulnerabilities, inconvenience with biometric systems, and imprecision with passive factors, leading to a need for a more reliable and user-friendly authentication method.
Innovation Solution
A system that authenticates and identifies users based on passive factors determined from sensor data, using a model trained with previous sensor data from the user's device to analyze feature vectors extracted from sensor data, such as movement-related data from accelerometers, to generate a security score and confidence value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If knowledge-based factors such as passwords are used for authentication, then users can be identified, but security is compromised due to password phishing, theft, and complexity requirements
Solution Approach 1:
The patent replaces knowledge-based authentication (passwords) with passive biometric authentication using sensor data collection. The system uses accelerometers, gyroscopes, and other sensors to capture movement patterns, device handling characteristics, and environmental interactions that uniquely identify users without requiring them to remember or manage passwords, thereby eliminating password vulnerability to phishing and theft.
Solution Approach 2:
The authentication system operates autonomously by continuously collecting sensor data in the background without requiring user awareness or action. The system automatically analyzes movement patterns and device characteristics to authenticate users, eliminating the need for users to manually provide authentication information and reducing human error in password management.
2Reliability
If biometric factors such as fingerprints are used for authentication, then authentication accuracy is improved, but user convenience deteriorates due to specialized hardware requirements and difficulty in altering biometrics if compromised
Solution Approach 1:
The patent employs a multi-sensor approach using accelerometers, gyroscopes, magnetometers, and other available sensors on standard mobile devices to create a comprehensive authentication system. This universal approach works on any device with basic sensors, eliminating the need for specialized biometric hardware while maintaining high authentication accuracy through combined passive factors.
Solution Approach 2:
The system creates dynamic authentication profiles by continuously collecting and analyzing sensor data that captures user behavior patterns, device handling characteristics, and environmental interactions. These dynamic profiles can be updated and adapted over time, allowing the system to maintain accuracy while accommodating changes in user behavior without requiring physical biometric alterations.
3Ease of operation
If passive factors such as cookies and IP addresses are used for authentication, then user experience is improved by eliminating additional user actions, but authentication precision deteriorates because such factors can only separate users into large classes
Solution Approach 1:
The patent transitions from traditional passive authentication factors (cookies, IP addresses) that operate at a coarse granularity level to a multi-dimensional sensor data space that captures fine-grained movement patterns, device handling characteristics, and environmental interactions. This dimensional expansion enables precise user identification while maintaining the background operation characteristic of passive authentication.
Solution Approach 2:
The system segments the authentication task into multiple independent sensor data collection and analysis components, each capturing specific aspects of user behavior. By analyzing acceleration patterns, rotational movements, device orientation changes, and environmental context separately and then combining them, the system achieves precise user identification through segmented analysis of multiple passive factors.
Data Source
AI summary
The disclosed embodiments relate to a system that authenticates and/or identifies a user of an electronic device based on passive factors, which do not require conscious user actions. During operation of the system, in response to detecting a trigger event, the system collects sensor data from one or more sensors in the electronic device. Next, the system extracts a feature vector from the sensor data. The system then analyzes the feature vector to authenticate and/or identify the user, wherein the feature vector is analyzed using a model trained with sensor data previously obtained from the electronic device while the user was operating the electronic device.