Passive Authentication via Siamese LSTM Sensor Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current user authentication methods on smartphones are cumbersome and pose security and privacy concerns due to the need for explicit user interaction, which can be vulnerable to social engineering hacks and biometric data theft.
Innovation Solution
A passive authentication system that uses a Siamese Long Short-Term Memory (LSTM) architecture to analyze data from multiple sensors on a user's device, such as keystroke dynamics and GPS location, to authenticate users without explicit interaction by comparing sensor data to pre-verified values and determining authentication decisions based on distance thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If explicit user interaction authentication is used (typing passcode or presenting biometric cue), then authentication reliability is improved, but ease of operation deteriorates and security vulnerabilities increase
Solution Approach 1:
The system performs authentication automatically without requiring explicit user interaction. The device monitors sensor data from the user's natural interactions (typing, swiping, holding) and autonomously determines authentication decisions, eliminating the need for users to actively participate in the authentication process while maintaining security.
Solution Approach 2:
The system collects and analyzes sensor data during the user's natural interaction with the device, performing authentication checks before critical operations are executed. This preliminary monitoring of interaction patterns allows the system to verify user identity proactively rather than reactively.
2Reliability
If explicit user interaction authentication is used, then authentication reliability is improved, but security concerns worsen due to vulnerability to social engineering hacks and biometric data theft
Solution Approach 1:
The system uses sensor data from natural interactions as an intermediary to verify user identity without directly handling sensitive biometric data or passwords. Instead of relying on traditional authentication credentials that can be stolen or phished, the system analyzes intermediate behavioral signals (typing rhythm, device handling patterns) that are difficult to replicate or steal.
Solution Approach 2:
The patent replaces traditional mechanical authentication systems (keyboards for passcodes, biometric scanners for fingerprints/faces) with a sensor-based system that monitors natural interaction patterns. This substitution eliminates the security vulnerabilities associated with traditional authentication methods while maintaining reliability.
3Reliability
If passive authentication with multiple sensor modalities is implemented, then security is improved, but device complexity increases
Solution Approach 1:
The system uses existing sensors in the device (accelerometer, gyroscope, touchscreen, microphone) for multiple purposes - both their original functions and authentication monitoring. This multi-functionality approach allows the system to gather authentication-relevant data without adding dedicated hardware, thereby limiting the increase in device complexity.
Solution Approach 2:
The patent combines data from multiple sensor modalities into a unified authentication decision process. Instead of treating each sensor separately, the system merges interaction patterns from various sensors to create a comprehensive behavioral profile, reducing overall system complexity through integration while maintaining high security.
Data Source
AI summary
A passive authentication method includes, in response to receiving a requested action from a first user, obtaining a set of sensor data and categorizing first sensor data of the set of sensor data into a first modality of a set of modalities. The method includes, for the first modality of the set of modalities, determining a distance value by applying a first modality model to the first sensor data and comparing the distance value to a first verified value of the first user for the first modality. The method includes, based on the comparison, determining a first authentication decision of the distance value. The method includes, in response to the first authentication decision indicating the first sensor data corresponds to the first user, performing the requested action.