Multi-granular User Behavior Profiles for Mobile Device Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobile device security methods, such as One Time Authentication (OTA) and Interaction-Based Authentication (IBA), are vulnerable to credential theft and mimicry attacks, which can allow unauthorized access.
Innovation Solution
A multi-granular authentication system that generates both long-term and short-term user behavior profiles, using Profile-Based Authentication (PBA) in conjunction with OTA and IBA, to monitor and compare user interactions, triggering strong authentication procedures when deviations from expected behavior patterns are detected, even in sleep states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If One Time Authentication (OTA) or Interaction-Based Authentication (IBA) is used, then authentication can be performed, but the credentials can be stolen and used to unlock the device or the device can be stolen after user has already logged in
Solution Approach 1:
The authentication system is segmented into multiple independent components: long-term behavior profiles (capturing days/weeks/months of usage patterns), short-term behavior profiles (capturing hours/days of usage patterns), and real-time authentication checks. This segmentation ensures that compromising one component does not compromise the entire authentication system, as each layer provides independent security verification.
Solution Approach 2:
Behavior profiles are generated and stored in advance through continuous monitoring of user interactions during training periods. The long-term profile captures extensive usage patterns over days, weeks, or months, while the short-term profile captures more recent behavior. This preliminary preparation enables rapid authentication decisions without requiring real-time complex analysis, and ensures authentication security is established before potential credential theft occurs.
2Measurement precision
If IBA techniques analyze accelerometer data patterns and swipe patterns to identify authorized user, then user identification can be performed, but the system is vulnerable to template theft and mimicry attacks
Solution Approach 1:
The system transitions from analyzing single-dimension behavior patterns (such as only swipe patterns or only accelerometer data) to multi-dimensional behavior analysis by capturing diverse interaction types including taps, scrolls, typing patterns, app usage sequences, and temporal patterns. This dimensional expansion makes mimicry attacks significantly more difficult as an attacker would need to replicate multiple independent behavior dimensions simultaneously, rather than a single pattern.
Solution Approach 2:
The authentication system implements nested profiles where the short-term behavior profile is contained within and supplemented by the long-term behavior profile. The long-term profile provides the foundational behavior patterns captured over extended periods, while the short-term profile provides more recent behavior context. This nested structure creates layered verification where both profiles must be satisfied, making template theft and mimicry attacks ineffective as attackers cannot replicate the nested hierarchical relationship between long-term and short-term behavior patterns.
3Reliability
If multiple profiles and monitoring mechanisms are implemented, then authentication security is enhanced, but the system complexity increases
Solution Approach 1:
The system dynamically adjusts the weight and importance of different behavior profiles based on current context and security requirements. The long-term profile and short-term profile can be selectively applied depending on the authentication scenario, and their influence on authentication decisions can be dynamically adjusted. This dynamic approach allows the complex multi-profile system to adapt to different security contexts without requiring static complex configuration for every possible scenario.
Solution Approach 2:
The system automatically performs profile generation, updating, and comparison without requiring manual intervention. During training periods, the system self-generates behavior profiles by monitoring user interactions and automatically updates them as new data becomes available. The authentication process automatically compares current behavior against stored profiles and makes authentication decisions without human involvement, reducing the operational complexity of managing multiple profiles while maintaining high security standards.
Data Source
AI summary
Techniques for authenticating a user of a mobile device at a computing platform are provided. A method according to these techniques includes generating a first profile and second profile of user behavior for the user of the mobile device, the first profile comprising a first type of profile having at least a first duration and the second profile comprising a second type of profile having a second duration that is shorter than the first duration, monitoring user behavior to generate usage behavior data, comparing the usage behavior data to the first profile and the second profile, performing a first type of authentication action responsive to the usage behavior data deviating from the first profile, and performing a second type of authentication action responsive to the usage behavior data deviating from the second profile.